a) By definition, Intel's implementation must be different than IBM's or anyone elses' because the CPUs aren't implemented the same. Not only do they implement different ISAs, but the entire architectures are different... different number of registers, different ISA, different designs.
D) This paper http://domino.watson.ibm.com/acas/w3www_acas.nsf/i...$FILE/heinrich.pdf , found on IBM's site by searching, is entitled Scalable "Multi-threaded Multiprocessor Architectures". The first paragraph states: "The former [hardware multi-threading], in the form of hyper-threading (HT) or simultaneous multi-threading, appears in the Intel Xeon and Pentium 4, and the IBM POWER5."
Well first off, I am not going to do everyone's homework on this, the info is out there, you all have Google. If you ask a IBM engineer if what Intel is doing is the same as what they are doing, or even if it is really SMT, they would tell you flat out that it is not and they fullfill completely different needs in their products and are implemented completely different. Your definition seems to be that the hardware can accept two threads, therefore it is SMT. That is a VERY simplisitic definition of what SMT is, when there are actually many variations on the concept(HT is a variation, but it is not what most CPU engineers consider actual SMT).
One of the primary issues here is that HT does not actually allow two simultanious threads, it is more of a enhanced thread scheduler that attempts to fill unused units with jobs that are pending. A true SMT CPU is actually architecturally able to execute two simultanious threads, its not just filling in idle parts of the pipeline with something to do(highly parallel designs). There is a ton of info on this, if you care I suggest you do the research yourself, I don't have the time(and in some ways the expertise) to write a lengthy article on the topic.
Alternatly, you can just buy into the marketing I suppose, its no skin off my teeth.
I was going to comment on the phrase "true SMT" above. I'm wondering if this comes from the same lines of thought as the "true dual-core" arguments.
Anyway, "HyperThreading" (HT) is just Intel marketing terminology for Symmetric MultiThreading (SMT). They are one and the same, with the same design goals... to more effectively utilize core resources by keeping the resources more busy instead of sitting around idle, particularly at the time granularity of cache misses and/or latencies.
#93 "Intel has labeled it as SMT, however there is another name for what they are doing(that I cannot remember at the moment). What they are calling SMT is nowhere even close to solutions like Power."
Well please tell us the exact definition of SMT and the difference between the multithreading in Power and P4?
"That aside, the implementation Intel has chosen is designed to make up for inefficiencies in the Prescott pipeline"
In Prescott pipeline? Why did the HT exist in Northwood based Xeons then? Of course the SMT is designed to reduce inefficiencies in the pipeline. If the CPU can utilize most of its resources when running a single thread there isn't a point of implementing SMT.
#93: Intel labeled SMT Hyperthreading. It is effectively the same as what the newer Power processors do (make one core two threads wide).
It also was not designed for Prescott, rather it was included in the P7 core from the beginning. For this reason it was available on P4s prior to Prescott.
HT improves the utilization of execution resources. Its not a bandaid, its a design choice. In some cases it can be used to compensate for some other weakness, in others it can simply be to increase throughput on multithreaded workloads.
Sun and IBM use it because they build server systems and SMT makes a large difference in traditional server loads.
Intel uses it because they realized it would work well with the P4. I don't know why AMD does not use it. Probably because they don't think the Athlon has enough unused hardware on typical loads to justify the extra transistors. Or maybe just because the Athlon was not designed with it in mind and they can't justify redoing the whole thing to add a single feature. Or maybe a combination of the two.
#93: Intel has labeled it as SMT, however there is another name for what they are doing(that I cannot remember at the moment). What they are calling SMT is nowhere even close to solutions like Power.
That aside, the implementation Intel has chosen is designed to make up for inefficiencies in the Prescott pipeline, such a implementation would make zero sense on the Athlon architecture, it does not share the same inefficiencies that the P4 design has. It would actually harm rather than help performance.
True SMT is not a 'bolt on' feature. Its something that has to be planned for from the very beginning of the CPU design cycle. You could not in any way add it to the current Athlon design and gain any performance. Whatever their next generation is may include it, it depends on what direction they decide to go, but you will not see it on the current generation, and thats actually a good thing as it would be purely a marketing move.
As always a very interesting article, one thing comes to mind though... In the gaming multitasking tests you adjusted the priority of the DVD Shrink process to see the effect on gaming performance. What I was wondering is if you could take a look at what effect explicitly binding the processes to seperate cores (processor affinity) has on gaming performance
Hyperthreading IS SMT. SMT stands for symmetric multithreading (ability to run two or more threads at once and this is exactly what hyperthreading does.
Of course, CPUs from different manufacturers have vastly different internal structures, thus also the SMT is implemented differently.
"Intel's next major IA-32 processor release, codenamed Prescott, will include a feature called simultaneous multithreading (SMT)"
to get repeatable multi-tasking/ncq benches, anand is going to have to bite the bullet and setup a full-blown network simulation:
1: an nntp server
2. a bittorrent swarm
3. an irc server
with this setup, you can test these multi-tasking scenarios that seem more reasonable:
1. firewall (a pig like zonealarm)
2. pulling news articles with either 2 clients or 1 client with 2 threads (writing to different places on hd simultaneously)
3. about 10 torrents where it is BOTH downloading and uploading (so pulling from a gazillion different places on hd at once)
4. mirc with about 5 open channels and some scripts (like filters). At least one channel should be very high traffic (like #mp3passion on undernet)
5. icq
6. running all this with software raid 5
this would represent a typical background load, and then you can benchmark foreground tasks to see how much they are affected by what's going on in the background (specifically ncq could be tested by seeing how long it takes to copy a file from one partition to another under these circumstances)
Just to be clear: SMT is NOT the same thing as HyperThreading. They go about what they are doing in radically different ways. The only similarity is in the CPU being able to execute two simultanious threads. How it goes about that though is implemented completely differently.
"#47, if HT is simply a "bandaid", then why is AMD the only major CPU vendor not using it? IBM uses it heavily in their Power5, Sun is making their next CPUs (Niagra) very highly SMT (same thing as HT). Arguably, both of those architectures have much more shallow pipelines than the P4, yet see reason to provide SMT. AMD is the only holdout."
The SMT used in IBM's Power series is completely different from what Intel is doing with the P4 design. The only similarity is the fact that two threads can be run at once, the implementation has nothing even close to the same however. I do not have details on Sun's implementation, but I would assume it will be closer to IBM's than Intel's implementation considering the market they are targetting. The Power architecture was designed from the ground up to use SMT, it wasn't a tacked on feature, and you get considerably more of a performance boost in most scenerios with it than you would ever see with HT on Intel.
The Athlon64 architecture was not designed with SMT or HT in mind, it was designed around two physical cores. So adding HT to it would do very little, and SSE3(which mostly optimizes HT style multithreading) does almost nothing on the K8 architecture.
Not every feature would help every CPU design, it all depends on what was taken into account when the design was made. Power has some limitations you do not see on x86(in order execution for example), and x86 has challenges you do not see on Power. The multi-threading implementations are similarly different and not comparable. In the x86 world, HT makes sense on Intel in some situations(not always). It makes no sense on AMD and would likely result in performance drops rather than gains. It certainly would not improve performance in any way as the core does not often have idle units or execution steps due to its design.
So you'd rather wait for information than recieve it now?
Anand clearly shows that dual core is only a good choice now IF you use it in scenarios where it can run multiple applications. Otherwise, single core chips are still the better choice. So i don't see the marketing hype you are referring to. Basically we've been told now for quite a few years that multi-taskers can benefit from multiple CPUs but the costs have been prohibitive. Now it looks like within the next year a 2 CPU machine will cost no more than previous single core processors.
Thanks Anand for helping us out in planning for the future! The DVDShrink stuff was very interesting to me as was the NCQ information - makes switching to SATA drives a bit more appealing to me considering my usage profile.
I just recently went from 1.4 k7->3.2 P4 w/HT so I'm pretty happy at this point. It does look like a dual core system *might* allow me to get rid of my second box (the 1.4 K7) which would save me money in the long run - one less PC to power up and cool off. My home office requires year round A/C to cool my 2 21" CRTs and 2 PCs...
I'm also curious to see what effects RAID would have on testing striped setups. I'm very curious if a RAID5 type of setup with NCQ and a dual-core might make chores like encoding & gaming realistic - it sounds like from the review that at this point I/O may cause hiccups even when the processor still has headroom.
"Make no mistake, Intel isn't officially releasing their dual core desktop processors today; this is merely a preview. Intel's dual core line is still on track to be released sometime in the April - June timeframe."
IF it where to be released in April then there should have been more reviews (sorry, previews).
IF it comes out in May then there is no need to do a preview this early (except to gain marketing hype).
IF it comes out in June then a "preview" this early should have contained technical information about the chip with a "review" containing benchmarks at a later date closer to its release date.
#81, Intel publicly stated that implementing HT on the P4 was about a 5% increase in logic/area on the chip. While it may be "ugly" and such, 5% certainly isn't that much of area when you consider how much area other functionality takes up.
#82, I haven't seen anyone announce duel core products. Dual core products, however, should be out very soon from Intel (and are [p]reviewed in the article this discussion is attached) for example). ;)
I must say i really can't relate to any of these multitasking scenorios. Well, except the teamspeak-gaming part.
What would be most interesting for me is dual-core, dual monitor setup. That would ROCK, and i'm really suprised noone remembered that. For instance, playing a divx on 1 monitor while gaming on other. Or watching TV on 1 while gaming on other (i do that a lot, if the football match is boring). These tasks are pretty much repeatable and can be measured. I was thinking about getting another monitor for these kind of purposes, but i'm not sure on how would my comp react to that kind of stress.
My personal thoughts about publishing an article on a duel core that is not yet availabile, nor will be for awhile, is that it is strictly marketing hype similar to what happened with the 1.13 GHz P3 or any other product that we the consumer have to wait for.
The idea that duel core would have a benifit in multitasking or multithreaded apps could have easily been demonstrated on a multiprocessor system using Opterons or Xeons. So now we get a review full of new (wow factor) benchmarks describing the new found benifit of having two processors working together.
I have no real problem with "preview" style articles, but having a plethora of benchmaks in a "preview" is pointless unless the goal is to advertise (hype) a product. The only people I see having any use for this "perview" are the ones that desperately want proof that their manufacturer of choice (this time being Intel) is superior, or those people who are so fickle that they will support the newest thing on the block no matter if it is needed, affordable or even the best option at the time it becomes available.
#47, if HT is simply a "bandaid", then why is AMD the only major CPU vendor not using it? IBM uses it heavily in their Power5, Sun is making their next CPUs (Niagra) very highly SMT (same thing as HT). Arguably, both of those architectures have much more shallow pipelines than the P4, yet see reason to provide SMT. AMD is the only holdout.
Seagate Barracuda 7200.7 NCQ - 25.2 minutes
Seagate Barracuda 7200.7 no NCQ - 33.6 minutes
Western Digital Raptor WD740 - 30.9 minutes
I'm shocked that the WD740 is slower. Isn't this the NCQ enabled 10,000 rpm drive? I would have expected this to beat out the 7200.7 NCQ seagate. Any reason why the WD740 would be slower?
Another point / comment:
Is it possible that DVD Shrink is slower on the AMD's because it has not been compiled / optimized for the AMD? Would this be any different on the 64 bit version of XP if DVD Shrink was compiled for 64 bit? Of course this may be moot as DVD Shrink is not longer being developed by the author.
Why do you still use the nForce 4 reference board for AMD tests? The real nForce 4 boards exist since over a quarter now and perform 10% better than the reference board. You also pretend to compare CPUs with similar price. Wouldn't it make more sense to compare combinations of mobo/CPU/memory of similar price?
This is not a fair comparison (Or is it the intention?)
In this atricle, I was unable to find cache information about the AMD chip (usually equipped with a 512 KB L2 cache).
To my point of view, in multiple threads scenario, the cache size may have dramatic influence and may explain several handicaps in the benchmarks. It would be nice to know what happens with a 1MB L2 cache Athlon 64 to keep things equal.
AMD chip has shorter branch prediction lines and seem quite well equipped for multitasking in theory, so why is it appearing so weak?
Thanks for the minimum frame rate comparison! And maybe you should not use so many flash-heavy pages, especially considering that you already told us that Athlon64 is much slower in Flash than Pentium4...
Looks like Pentium D is a better choice in more ways than the Prescott is
The multitasking gaming analysis combined with HT and NCQ was very insightful, almost pioneering...
and with dual core AMD's people have to remember that the whole architecture is designed differently, and hence the possible suprises in tasks that the pentium D may not perform well on
but the power consumption advantage on amd is mighty tempting as amd will be the cpu that just keeps on saving -- with power bills that is
I was thinking, seeing most games are single-threaded you might like to try and benchmark this scenario on the dual core machine:
Set all the processes including operating system processes and other background processes to run on CPU 0 (first core). Then set the particular game to run on CPU 1 (second core). Have nothing else but the game running on CPU 1. This should dedicate a whole CPU core to the game for maximum performance in theory. I believe you can set the affinity in the task manager or use a batch file to do it.
Then run a Doom3 benchmark or HL2 benchmarks to see if the gaming performance is increased by letting the single-threaded game have a whole CPU core with no interuptions.
The AMD not competing with the PD 2.8 will be due to price.
Remember that AMD will need to sell a 2 core 2.2 GHz processor for about double the price of the single core version (maybe even more if yields are a problem). Therefore, you should expect a dual core 2.2 proc to be closer to $500-600 instead of the ~$300 processors tested here.
This also explains why AMD is focusing on the workstation market first. It's an area where price is typically not as much a factor, and where they are competing with Xeon prices, so it will be much easier for them to sell procs @ the higher prices.
I'm guessing that AMD's desktop dual core procs will start @ 1.6 or 1.8 GHz, and be priced on par with the Intel offerings. I think the overall performance will be similar to the current single core behavior, but now you can have more stuff running.
I currently have an AMD 939 3200+, so I am looking forward to their dual core offerings in a couple years, and hoping my mobo will be compatible. That would be very cool.
Also, I would like to add vote for a WoW test with Teamspeak running in the background as well as an istance if IE or Firefox with tthotbot.
Any chance you could give us a little peak at what dual opterons on the nforce pro chipset can do in the same benches that you did in this article.
A configuration something like this:
Opteron 252s clocked at 1.8GHz, 2.2GHz, 2.6GHz
1 GB ECC DDR400
NVIDIA nForce pro motherboard
ATI Radeon X850 XT PCI Express
NCQ enabled HDD (maxline III, 7200.8, etc)
I looked through all the old articles I could find, but most seemed to contain only server oriented benches. I think that kind of article would be very enlightening as to the future performance of dual core athlons (probably within a couple percent)... maybe help put to rest some of the questions a lot of us have about waiting for amd dualcore, intel dualcore or just overclocking the heck out of a venice a64.
SSE3 does almost nothing for the Venice which is barely faster than the Winchester(Xbit Labs benchmarked it already)
While Anandtech said the dual cored A64s wouldn't compete with the Pentium D in encoding, unless they actually have dual cored A64s which they can't show us, i'd be willing to argue with that. Encoding is one of those things thats largely affected by Hyper-Threading and the Pentium D loses hyper threading. When the A64 goes dual, its performance increase will be larger than Intels.
Last time I checked (which admittedly was a while ago), SMP support was broken in the later builds of Q3A. I can't remember if it was Quake 3 or the combination of Q3 and ATI/NV drivers, but the performance stopped improving.
It`s nice to finaly see some competition from Intel.
They slapped together theyr old stuff in a new package. But we all know that a new package isen`t going to change anything(Like wrapping s*** in gold paper).
Be happy as longe as it last, and have your 15min of fame.
Remember they rushed out the dual core, and they did it for you IndelJugen!.
Thanks for the article Anand...none of us take Charlie seriously anyway...
"AMD's dual core will be quite impressive, even more so than Intel's"
I am hearing the same. There is some serious research work being done in the TV and Film industry right now with the dualcore Opterons, and it is MOST impressive! Still under NDA (as are we all), I can only say that the results so far have been much better than expected!
Thank You Anand for the great article, especially the info on the NCQ. Great writing, and overall a very good read.
btw, I understand how fusterated you must feel making these benchmarks, not having things work, trying to remember all the things you want/have to do next, ect......Keep it up Anand, that is why you are the best!...try to get some sleep though m8 :-P
I would probably say that the 2.2 ghz from AMD it would not be compared to the dualCPU in this article because if the 2.2 is going to probably be the FX line, then it would be compared with the top of the line of Intel's...remember this was an article about "value" dual cores (oxymoron ;-))...so due to price and probably performace it would not be paired with the Pentium D at 2.8...sort of like AMD's naming scheme, an AMD 2800 at 1.8 ghz matches up with a 2.8 ghz Intel...so I would assume that AMD's biggest baddest dualy will blow Intel out of the water...and not because i am an incessant AMD fanboy because i am an avid gamer, but becuase of AMD's past performace, and AMD architechure is designed for dual core. We shall see, wh shall see...
*STATEMENT: The author of this post is not hereby responsible for any grammatical errors, typing, or syntax, of any kind.* lol
In the future I cannot Imagine the power requrements of CPU's they may end up needing their own 500watt dedicated supply and a second one for HDD's, GPU's, Fans, motherboards, and accessories.
The only thing more ineffieciant than a 250watt fully loaded Prescott is the old eniac, It was said that when it was turned on the Whole City of Philadelphia would go in to a brown out. I am afraid that modern processors are taking steps back instead of forward.
Hmm... if permormance isn't the only metric for dual-core AMD, then perhaps power also plays a role? What if they could craft a 2.2gHz monster that had the same power draw as a 2.0gHz a64? Such a beast, besides being a processing power-house, would show extreme overclock potential.
'The move down to 90nm really cut down AMD’s power consumption a lot, to the point where the 90nm Athlon 64 3500+ actually consumes less power under full load than the Pentium 4 630 at idle.'
ROFL, I expect the old Prescott space heater gags to start again after this review :)
on page 13 top diagramm it says "Thanks to DVD Shrink behaving and running with a low priority, our gameplay was largely unaffected on the Athlon 64. The performance dropped less than 3% in Doom 3. ..." yet the graph shows the A64 at the bottom far behind the Pentiums. Is the graph messed up or the comment?
Great article though and thatnk you for listening to your readers.
Thanks also for the NCQ-page. I still wonder if you want to pick up where you left off some months ago and look into RAID in more detail, possible measuring the impact of NCQ on the benefits of SMT or vice versa and CPU-load issues with simple RAID5 setups on dual-core?
Let's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz.
AMD's dual core will be quite impressive, even more so than Intel's. Don't look at performance as the only vector to measure though...
....PRICE.
I think you were spot on. 2.2GHz DC Athlon is almost certain NOT to compare to PD2.8GHz.
The article is comparing 3 systems with the same price processor.
The comments from Anand suggest that the 2.2GHz AMD DC will blow the 2.8GHz PD away, but at a higher price.
The 1.8GHz AMD DC is more likely to be comparable (and possible faster, if the 2.2GHz Athlon really is amazing). But who knows?
Rumors indicate AMD's aiming for a launch at 1.8GHZ - 2.2GHZ frequencies for the Athlon 64 Dual Cores, I would guess, that most likely that these Dual Cores are based on the San Diego cores, as they each have access to 1MB of cache with a die surface area slightly larger then Clawhammer.
Another guess would be they would use the 2.2GHZ Dual Cores as their FX flagship and have the 1.8GHZ/2.0GHZ variants as their mainstream line.
However with AMD saying 2H 2005 for their Dual Cores, we won't have to wait too long then for Intel's new version of Dual Core in 1H 2006 on the 65nm process.
Maybe the 2.2GHZ Athlon 64 Dual Core is much more expensive then the 2.8GHZ Pentium D? Or it competes with Intel's Pentium D 3.2GHZ.
Great review! It is refreshing to see different kinds of benchmarks.
The problem with dual core Athlon64 is that it won't be launched until 2006 according to the AMD's roadmap: http://www.amd.com/us-en/Processors/ProductInforma... (dual core Athlon64 FX will be launched in H2 2005, but no regular dual core Athlon64 is scheduled for 2005)
And in 2006, Intel will have 65nm dual core chips available...
For those saying that Intel will have ANOTHER dual core system out because of the delay by AMD...remember...this was a technology PREVIEW. There is no released hardware yet.
Its going to be interesting to see what the benefits of AMD's built from the ground up to be multi-core approach vs. Intel's patch job (I still have utmost respect for Intel, they created one fine "patch job" under a severe time constraint, sometimes it does pay to be a behemoth ;)) will be.
Don't take this post the wrong way, I am not knocking Intel's HyperThreading, but I think most people have actually forgotten what it is. It is not a feature of the Pentium 4, but a bandaid approach to keep their now 31 stage pipeline busy doing something. The reason that a 3.8GHz P4 doesn't run circles around a 2.6GHz A64 is that horribly designed 31 stage pipeline. AMD can't implement some form of HyperThreading because it doesn't have a bunch of processor stages sitting around doing nothing like a P4 does without HyperThreading.
#45 - The Athlon 64 still lags behind on encoding cabability and its been shown over the past year. However some of the content tested was designed specifically for P4 optimization so its hard to get a reliable result. Who knows... until we see the new AMD64 core with SSE3 we cant really pass judgement.
The first graph on the DVD Shrink/Game bench still seems a bit suspect. Why would the P4 outperform the PD and the A64 when under normal circumstances the A64 should be superior in singlethread and the PD in multithread?
Anway, great article. I'm really looking forward to seeing what AMD has to offer since while I'd greatly like improved multitasking I'd also like a good gaming CPU.
I have to ask then, is it because of AMD's onboard memory controller that they don't have to manufacture another motherboard for the dual core CPUs? If so, you think AMD was thinking about this scenero (dual cores) well before the first clawhammer came out?
That would sound impressive. Compared to Intel's dual press hots.
#37 - Well im hard pressed to really want one when my current setup is still sufficient. That and I have my heart set on AMD64 for gaming. Even at DVD backup I can do a movie in about an hour even with full quality under DVD Shrink. And really, I would use DVD Rebuilder which is very much single threaded and in my book Quality > Speed. Takes about 6-7 hours but better results than DVD Shrink could have wished for.
Hey Anand, I noted that you said in Part one that the Intel Dual core was one of the most stable config's you ever had. However in part two and quote, "After recovering from Part I and realizing that my nForce4 Intel Edition platform had died, I was hard at work on Part II of the dual core story. "
Was this human error or was it a manufacturing error?
#26: Actually, encoding a DVD should be a multi-tasking event in of itself!
Task1: DVD creation; menu's, transitions, etc
Task2: DVD encoding; background rendering of menu's and transitions
Task3: DVD encoding; background rendering of the actual menu
Task4: Burning of DVD; you should be able to start burning the DVD before the encoding in task 2 or 3 is complete, as long as the burner is properly buffered
All i know is that i alt-tab / alt-enter to the desktop running general apps all the time while gaming. I bought two system so i can download while gaming on the other system for this very reason. To do both at the same time would cause the ftp software to go into idle state with the fastest download speed at only 8-10kb/s. I can set the ftp software at a higher priority but then it would just cripple my gaming. These dual core look very promising, but ill hold out for amd dual core.
I'm interested in benchmarks that would be relevant to scientific computing and software development.
How about benchmarking a parallel compile of some non-trivial software package such as building the
gcc compiler. That takes quite a long time on my 4 year old laptop.
What about the fourth option? That by the time AMD's desktop dual core processor is available, Intel will have a new dual core processor available. Now, whether we're talking more than "two cores bunged on a chip", or whether we're simply talking a jacked up FSB (which has, remember, always given Intel a hefty jump in the encoding arena), I don't know. But I do know one thing...
Intel is a very big company and Intel has very big sleeves. ;)
He said that a 2.2 GHz dualcore Athlon 64 wouldn't compete with the 2.8 GHz Pentium D at encoding. Notice the encoding part, he said nothing about other stuff.
I'm guessing one can know that by looking at dual CPU Opteron systems, the dualcore A64 won't beat them, and if they can't beat a 2.8 Pentium D then the dualcore A64 won't be able to either.
Is there something wrong with the graphs in the DVD Shrink/Game test? The comments doesn't seem to match them (especialy the part about the minimum frame rate being equal)
How do you calculate the system wattages like that? I have been attempting to find detailed information on how to do this but have turned up nothing yet.
#10 - you are quite correct. anyone who games with a processor-intensive background task running at the same time _on a single core processor_ is insane. the reason I wanted to see benchmarks is to see whether dual core changes that.
theoretically, I don't see why it wouldn't work:
- you only have one GPU, and only the game is using it
- you have two processor intensive tasks - the game and the background task, and two cores, one for each
hence, no conflict. whether that actually holds up in the real world is/was the question (if the background task is multithreaded, or heavily uses reasources other than just the processor, then naturally the above doesn't hold true).
Since I mostly play games ill stick to buying the AMD64 3500+. Thanks. My definition of multi-tasking is using a whole other computer ;)
The Pentium D seems pretty decent at multitasking as you would define running two things at once but I rarely do that sort of thing since its kind of dumb to encode a DVD in the background while playing a game. Or does encoding a DVD really interfere with browsing the web? I dont know... that and the heat factory output as if it was bad enough is now worse.
#23 - I'm assuming that a dual core A64 at 2.2GHz will blow a Pentium D out of the water at any of the launch frequencies! The Prescott core isn't really designed for multi-core operation, and needs some kind of arbitration logic and some funky-memory-controlling-thingy to work. As a result, the performance improvements in multi-threaded applications aren't anything like the theoretical extra 100% another core could bring. With A64 being designed for multi-core operation I'd expect the increase in performance to be nearer 85%.
As regards the performance gap between the P4 630 and A64 3500+, the majority of the benchmarks shown here are designed specifically to show performance improvements in multi-core processors. The 630 is hyper-threaded and therefore logically multi-cored, if not physically so. As such the 630 will have artificially high performance compared with the 3500+ - in most single-application benchmarks, the AMD chip would thrash both Intel chips.
Is there any chance of adding benches for the 630 with HT disabled (or at least giving us an idea of performance.)? We've got a vanilla A64 versus a HT 630 and Dual-core system. It'd be good to see how a single core performs for reference.
I really can't say more, but you are barking up the wrong tree with those assumptions :)
AMD's dual core will be quite impressive, even more so than Intel's. Don't look at performance as the only vector to measure though...
marcusgarcia
We did look at HT performance when it came out, but the problem is that HT doesn't improve performance in all cases. Look at the Gaming Multitasking Scenario 2 tests, HT reduced performance significantly - most likely because DVD Shrink and Doom 3/Splinter Cell were both contending for floating point resources that were in use. Dual core solves this problem by having two complete sets of execution units, so there's no worry about contention between threads for shared resources.
As far as Half Life 2 goes, it is still single threaded so its performance characteristics would be no different than what you see here.
mlittl3
I've been looking into running VoIP or some sort of voice chat program in the background, the problem surfaces in trying to put together a reliable, repeatable workload. Dual core will most definitely help there, but how much - I do not know.
I haven't given up yet :)
BruceDickenson
Glad to have you on-board and thanks for the kind words :)
Woodchuck2000
The new dual core chips are still LGA-775, but they do require a new motherboard (unlike AMD's solution which just requires a new BIOS). Currently Intel's 945 and 955 chipsets will support dual core, and tomorrow I should have a nForce4 SLI Intel Edition board that will support dual core as well. The new NVIDIA chipset does support dual core, but it's up to motherboard makers to implement support for it in their designs.
Check with the motherboard maker to make sure that dual core is explicitly supported by the board, it should say so somewhere in the manual or on the box.
#22... Anand said a 2.2 GHz Athlon 64 won't compete with the 2.8 GHz Pentium D. That either means a 2.2 GHz dual core Athlon 64 will have lackluster performance, or it will be AMD's new enthusiast line like the FX is right now, which means it would be competing with the Extreme Edition chips, not the regular line.
I guess there's a 3rd possibility. He was referring to dual core Opterons which obviously won't compete with the Pentium D any more than the Opteron competes with the Pentium 4 right now.
Just out of interest, does anyone know what socket do the new cores use? Will the new nVidia chipset support the new cores (it was hinted at briefly, but not stated explicitly...)?
#19 - What's your source for those assertions? I've heard reports that AMD have got samples running at well over 2GHz and since the K8 architecture is natrually better suited to multiple cores I'd have expected blistering performance. BTW, does anyone know if the AMD cores will be based on the new Venice rev? Is SSE3 a given?
Hey all, long time reader, this is my first "post"/comment...
Just had to say this is one of the most interesting articles I've read in a long time. I loved the NCQ tangent, it almost felt like you were part of a conversation when you read how Lal Shimpi analyzed the anomaly in his testing.
Forgot to say in the last post, my rant is about HT, not dual cores.
I know 2-cores won't make THAT difference on these trivial things (who needs another 2.8ghz for simple stuff?)...but..HT is benefiting GREATLY from it, yet noone mentioned it and didn't even try this sort of test when HT was launched.
When you see the 3.0 HT doing better than a AMD 3500+ (supposedly 500 points faster), you gotta ask how badly would it beat the AMD 64 3000+, which happens to cost almost the same than the P4 3.0 ghz...which happened to destroy the much faster AMD on the test.
That pretty much sucks and leave us with the impression that people either:
a - wanted to benefit AMD
or
b - were too ingenuous to think on these tests when doing HT tests (which can't be true because i always wanted them)
1.) The dual core 2.2 GHz Athlon-64's will be less than impressive and won't even perform in the same class as a Pentium D @ 2.8 GHz.
2.) The mainstream Athlon-64 dual core chips will be running at much less than 2.2 GHz, and the 2.2 GHz dual cores will be the FX line, which compete with the Extreme Edition Pentiums.
1º: Outlook checks 8 pop accounts for mail and apply it's rules to it every minute or so.
2º: MSN with webcam can eat quite some CPU, specially because i play on the dark with the "low light filter" turned on, which happens to eat quite some CPU.
3º: For every file opened/closed both the AVG and the MS anti-spyware are going to have a check if that's malicious and if the action is allowed.
When i close everything and run 3dmark01 i get around 300 - 600 more points out of it from my 12200 points score.
PS: don't forget IE, which is usually opened here or on tom's hardware (or both and some more), which happen to have a lot of those huge flash banners.
I think that DOES make *a lot* of difference.
Add to that the fact of many people using skype while gaming, mainly on FPS and RTS, which can make all the difference.
"Let's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz."
So you leave two possibilities.
1.) The dual core 2.2 GHz Athlon-64's will be less than impressive and won't even perform in the same class as a Pentium D @ 2.8 GHz, but rather dual core Extreme Edition chips.
2.) The desktop dual core Athlon-64's will be running at much less than 2.2 GHz.
A really cool multitasking scenario for gaming would be running a game with something like Skype in the background. Everyone saying that a respectable gamer (whatever that means) would not run multi applications in the background is not thinking about VoIP.
I am in Louisiana and I like to game with my friend in Georgia. We talk to each other using Skype will playing Halo on the same server. I know the overhead necessary for VoIP must slow things down some.
I know I personally have a lot of things open when gaming, especially if I'm playing World of Warcraft. I'll typically alt-tab out of the game to check IRC or Firefox (with a bunch of tabs open) to look something up or if I'm bored, just browse the net a little bit.
The only problem I ever have with slowdowns is if the game is highly CPU-bound and uses up 100% of my CPU, which WoW does almost all the time.
#10- I'm inclined to agree, but people did request it so presumably some people ar interested in doing so for whatever reason.
"I don´t close AVG and MS Antispyware and MSN and outlook and IExplorer everytime i open warcraft or half life 2, so...WHO MADE ME BELIEVE AMD WAS FASTER?"
Merely running applications in the background isn't going to do much to benefit DualCore/SMP unless those applications are actually utilizing the processor. Odds are MSN/Outlook/Spyware/Anti-virus probably aren't doing a thing but sitting idle when your gaming.
I play warcraft III online, which is a RTS game.
Being so, not all actions are dependant on my reflex (in fact i can many times minimize the game for around 20 seconds which is the time my char takes to walk to a certain place on a given map).
That being said, i am ALWAYS with a few instances of internet explorer open, MSN open, outlook express open and of course AVG and MS Anti spyware loaded on memory with real time protection.
Add the fact that sometimes i am viewing and being viewed on MSN webcam.
I'm sure MANY more people do that.
Remember not all players are FPS gamers...in fact, FPS is far behind MMORPG in sales, which doesn't require near as much attention and reflex.
I mean, WHY DIDN'T ALL SITES DO THESE TESTS WHEN HT WAS LAUNCHED?
It clearly shows here what is MUCH better when it comes under regular usage.
A Pentium 4 3.0 ghz is beating AMD's trash on 3500+....i mean, WTF?
Almost noone (does anyone at all?) goes closing all applications before gaming or doing any other activity and HT is clearly giving AMD a serious beating on the multi-tasking scenario (read: EVERYONE's usage).
I don´t close AVG and MS Antispyware and MSN and outlook and IExplorer everytime i open warcraft or half life 2, so...WHO MADE ME BELIEVE AMD WAS FASTER?
I mean...dude...are we talking about servers here to compare single threaded performance?
Are we still on Windows 3.11?
By the way, how in the hell aren't they including Half Life's 2 performance?
Surely the physics engine plays quite a bit on processing and even more surely it is done on separate threads, which would show the dual core being strong even on a single application, let alone on a multi-tasking one.
I'm quite repented for having an Athlon 64 3000+ as my CPU right now when the Pentium 4 3.0 HT would be clearly outspacing the Athlon in every respect as long as i was multi-tasking/opening/closing/minimizing things (e.g.: ALWAYS).
i think that this real-world multitasking testing done by Anandtech is 1000 times better than one syntetic benchmark, that is most time optimized for one or another cpu....
someone agree ?
I'm sorry but I find the premise of Page 11 borderline absurd. I *cannot* fathom there's a respectable amount of gamers that actually do that on a regular basis.
Multi-core multitasking is already quite difficult, you have no idea how frustrating last weekend was. The issue is that I can sit with you on a computer and show you all the areas that dual core will improve performance, but quantifying it so I can stick a bunch of bars in a graph is far more difficult. AMD and Intel are actively working with BAPCo on SYSMark 2006 that should be much more multi-core friendly, but until then we're left with a lot of hard work. We're trying to write our own benchmarks as well, it's just that they take quite a bit of time to put together.
Thanks for pointing out the graph error, the labels just got messed up it looks like; should be fixed now.
Remember AMD is talking about a 2H 2005 launch for dual core Athlon 64 on the desktop, don't expect to see reviews of desktop parts anytime soon.
As far as the encoding comment goes, it's tough for me to actually elaborate without stepping into areas I can't get into just yet. Let's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz.
Ok Anand, either you're slick or you're slipping. History shows you're slick..
You said a dual-core A64 won't help in encoding apps. I know you're not one to say stuff just because you THINK it's true, but because you KNOW so. I'm not at T0M H4rdware..
So.. Since you're alluding to it, WHEN'S THE DUAL CORE A64 TEST COMING OUT!?!?! *pant**pant*
Seriously though, I see that this multicore, multi-tasking benchmarking is going to get quite difficult. How do you know just how fast it really is considering all the combinations of different apps you will have running in the background? It Madonion or those other benchmarking guys going to be coming out with a synthetic benchmarking tool to gauge the max performance of these new multi-core processors?
hm, with a P4 EE or PDEE, with 2 6800Ultra or GT, you're definitely expecting an oven case.
Wonder how those dualcores overclock-->that should help to make up the single thread performance. On the other hand, Clearly HyperThreading is GOOD, but AMD still says NO.
The only thing I have to question is how a dual core Athlon-64 at encoding. On the last page you say, "For encoding performance, you still can’t beat the Pentium D. Even a dual core Athlon 64 isn’t going to help enough in that area."
What makes you think that when the 2.2 GHz Athlon-64 is RIGHT behind the 3.0 GHz Pentium 4 630?
Everything I've read so far has told me that the Athlon-64 should scale better than the Pentium 4 as far as dual cores are concerned... so... what exactly are you basing your opinion that even a dual core Athlon-64 won't help in encoding???
I guess it makes sense that NCQ would help when multitasking. I assume that this would be the same on single cores as well? The new focus on responsiveness is a good move IMO. The time to switch between apps and redraw the screen and clicking on menus and buttons is what frustrates me when I am multitasking.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
106 Comments
Back to Article
BoBOh - Monday, April 11, 2005 - link
Where are the code compile tests. We're not all gamers, some are software developers! :)BoB
NightCrawler - Saturday, April 9, 2005 - link
Dual core Athlon 64's in June ?fitten - Saturday, April 9, 2005 - link
- also, there should be (SMT) after simultaneous multi-threading in the quote from the paper on the IBM site.fitten - Saturday, April 9, 2005 - link
- quote should be in front of "Scalable not after.fitten - Saturday, April 9, 2005 - link
a) By definition, Intel's implementation must be different than IBM's or anyone elses' because the CPUs aren't implemented the same. Not only do they implement different ISAs, but the entire architectures are different... different number of registers, different ISA, different designs.2) Intel's definition of HyperThreading: http://www.intel.com/technology/hyperthread/
D) This paper http://domino.watson.ibm.com/acas/w3www_acas.nsf/i...$FILE/heinrich.pdf , found on IBM's site by searching, is entitled Scalable "Multi-threaded Multiprocessor Architectures". The first paragraph states: "The former [hardware multi-threading], in the form of hyper-threading (HT) or simultaneous multi-threading, appears in the Intel Xeon and Pentium 4, and the IBM POWER5."
Reflex - Friday, April 8, 2005 - link
Well first off, I am not going to do everyone's homework on this, the info is out there, you all have Google. If you ask a IBM engineer if what Intel is doing is the same as what they are doing, or even if it is really SMT, they would tell you flat out that it is not and they fullfill completely different needs in their products and are implemented completely different. Your definition seems to be that the hardware can accept two threads, therefore it is SMT. That is a VERY simplisitic definition of what SMT is, when there are actually many variations on the concept(HT is a variation, but it is not what most CPU engineers consider actual SMT).One of the primary issues here is that HT does not actually allow two simultanious threads, it is more of a enhanced thread scheduler that attempts to fill unused units with jobs that are pending. A true SMT CPU is actually architecturally able to execute two simultanious threads, its not just filling in idle parts of the pipeline with something to do(highly parallel designs). There is a ton of info on this, if you care I suggest you do the research yourself, I don't have the time(and in some ways the expertise) to write a lengthy article on the topic.
Alternatly, you can just buy into the marketing I suppose, its no skin off my teeth.
fitten - Friday, April 8, 2005 - link
I was going to comment on the phrase "true SMT" above. I'm wondering if this comes from the same lines of thought as the "true dual-core" arguments.Anyway, "HyperThreading" (HT) is just Intel marketing terminology for Symmetric MultiThreading (SMT). They are one and the same, with the same design goals... to more effectively utilize core resources by keeping the resources more busy instead of sitting around idle, particularly at the time granularity of cache misses and/or latencies.
defter - Friday, April 8, 2005 - link
#93 "Intel has labeled it as SMT, however there is another name for what they are doing(that I cannot remember at the moment). What they are calling SMT is nowhere even close to solutions like Power."Well please tell us the exact definition of SMT and the difference between the multithreading in Power and P4?
"That aside, the implementation Intel has chosen is designed to make up for inefficiencies in the Prescott pipeline"
In Prescott pipeline? Why did the HT exist in Northwood based Xeons then? Of course the SMT is designed to reduce inefficiencies in the pipeline. If the CPU can utilize most of its resources when running a single thread there isn't a point of implementing SMT.
saratoga - Friday, April 8, 2005 - link
#93: Intel labeled SMT Hyperthreading. It is effectively the same as what the newer Power processors do (make one core two threads wide).It also was not designed for Prescott, rather it was included in the P7 core from the beginning. For this reason it was available on P4s prior to Prescott.
saratoga - Friday, April 8, 2005 - link
#80:HT improves the utilization of execution resources. Its not a bandaid, its a design choice. In some cases it can be used to compensate for some other weakness, in others it can simply be to increase throughput on multithreaded workloads.
Sun and IBM use it because they build server systems and SMT makes a large difference in traditional server loads.
Intel uses it because they realized it would work well with the P4. I don't know why AMD does not use it. Probably because they don't think the Athlon has enough unused hardware on typical loads to justify the extra transistors. Or maybe just because the Athlon was not designed with it in mind and they can't justify redoing the whole thing to add a single feature. Or maybe a combination of the two.
saratoga - Friday, April 8, 2005 - link
#90:HT is the same thing as SMT. You can thank Intel's marketing for that one.
Reflex - Friday, April 8, 2005 - link
#93: Intel has labeled it as SMT, however there is another name for what they are doing(that I cannot remember at the moment). What they are calling SMT is nowhere even close to solutions like Power.That aside, the implementation Intel has chosen is designed to make up for inefficiencies in the Prescott pipeline, such a implementation would make zero sense on the Athlon architecture, it does not share the same inefficiencies that the P4 design has. It would actually harm rather than help performance.
True SMT is not a 'bolt on' feature. Its something that has to be planned for from the very beginning of the CPU design cycle. You could not in any way add it to the current Athlon design and gain any performance. Whatever their next generation is may include it, it depends on what direction they decide to go, but you will not see it on the current generation, and thats actually a good thing as it would be purely a marketing move.
eeceret - Friday, April 8, 2005 - link
As always a very interesting article, one thing comes to mind though... In the gaming multitasking tests you adjusted the priority of the DVD Shrink process to see the effect on gaming performance. What I was wondering is if you could take a look at what effect explicitly binding the processes to seperate cores (processor affinity) has on gaming performancedefter - Friday, April 8, 2005 - link
Hyperthreading IS SMT. SMT stands for symmetric multithreading (ability to run two or more threads at once and this is exactly what hyperthreading does.Of course, CPUs from different manufacturers have vastly different internal structures, thus also the SMT is implemented differently.
"Intel's next major IA-32 processor release, codenamed Prescott, will include a feature called simultaneous multithreading (SMT)"
http://arstechnica.com/articles/paedia/cpu/hyperth...
tynopik - Friday, April 8, 2005 - link
and of course that's just the net part, don't want to leave out other background tasks like that resource sucker outlook and playing flac/ape filestynopik - Thursday, April 7, 2005 - link
to get repeatable multi-tasking/ncq benches, anand is going to have to bite the bullet and setup a full-blown network simulation:1: an nntp server
2. a bittorrent swarm
3. an irc server
with this setup, you can test these multi-tasking scenarios that seem more reasonable:
1. firewall (a pig like zonealarm)
2. pulling news articles with either 2 clients or 1 client with 2 threads (writing to different places on hd simultaneously)
3. about 10 torrents where it is BOTH downloading and uploading (so pulling from a gazillion different places on hd at once)
4. mirc with about 5 open channels and some scripts (like filters). At least one channel should be very high traffic (like #mp3passion on undernet)
5. icq
6. running all this with software raid 5
this would represent a typical background load, and then you can benchmark foreground tasks to see how much they are affected by what's going on in the background (specifically ncq could be tested by seeing how long it takes to copy a file from one partition to another under these circumstances)
Reflex - Thursday, April 7, 2005 - link
Just to be clear: SMT is NOT the same thing as HyperThreading. They go about what they are doing in radically different ways. The only similarity is in the CPU being able to execute two simultanious threads. How it goes about that though is implemented completely differently.Reflex - Thursday, April 7, 2005 - link
"#47, if HT is simply a "bandaid", then why is AMD the only major CPU vendor not using it? IBM uses it heavily in their Power5, Sun is making their next CPUs (Niagra) very highly SMT (same thing as HT). Arguably, both of those architectures have much more shallow pipelines than the P4, yet see reason to provide SMT. AMD is the only holdout."The SMT used in IBM's Power series is completely different from what Intel is doing with the P4 design. The only similarity is the fact that two threads can be run at once, the implementation has nothing even close to the same however. I do not have details on Sun's implementation, but I would assume it will be closer to IBM's than Intel's implementation considering the market they are targetting. The Power architecture was designed from the ground up to use SMT, it wasn't a tacked on feature, and you get considerably more of a performance boost in most scenerios with it than you would ever see with HT on Intel.
The Athlon64 architecture was not designed with SMT or HT in mind, it was designed around two physical cores. So adding HT to it would do very little, and SSE3(which mostly optimizes HT style multithreading) does almost nothing on the K8 architecture.
Not every feature would help every CPU design, it all depends on what was taken into account when the design was made. Power has some limitations you do not see on x86(in order execution for example), and x86 has challenges you do not see on Power. The multi-threading implementations are similarly different and not comparable. In the x86 world, HT makes sense on Intel in some situations(not always). It makes no sense on AMD and would likely result in performance drops rather than gains. It certainly would not improve performance in any way as the core does not often have idle units or execution steps due to its design.
Icehawk - Thursday, April 7, 2005 - link
"I'm also curious to see what effects RAID would have on testing striped setups."Uh, delete the "striped setups" from the end ;)
Can we please, please get some kind of short-term editing abilities here?
Icehawk - Thursday, April 7, 2005 - link
So you'd rather wait for information than recieve it now?Anand clearly shows that dual core is only a good choice now IF you use it in scenarios where it can run multiple applications. Otherwise, single core chips are still the better choice. So i don't see the marketing hype you are referring to. Basically we've been told now for quite a few years that multi-taskers can benefit from multiple CPUs but the costs have been prohibitive. Now it looks like within the next year a 2 CPU machine will cost no more than previous single core processors.
Thanks Anand for helping us out in planning for the future! The DVDShrink stuff was very interesting to me as was the NCQ information - makes switching to SATA drives a bit more appealing to me considering my usage profile.
I just recently went from 1.4 k7->3.2 P4 w/HT so I'm pretty happy at this point. It does look like a dual core system *might* allow me to get rid of my second box (the 1.4 K7) which would save me money in the long run - one less PC to power up and cool off. My home office requires year round A/C to cool my 2 21" CRTs and 2 PCs...
I'm also curious to see what effects RAID would have on testing striped setups. I'm very curious if a RAID5 type of setup with NCQ and a dual-core might make chores like encoding & gaming realistic - it sounds like from the review that at this point I/O may cause hiccups even when the processor still has headroom.
justly - Thursday, April 7, 2005 - link
#84 What do you consider "very soon"?
The following was quoted from "part 1".
"Make no mistake, Intel isn't officially releasing their dual core desktop processors today; this is merely a preview. Intel's dual core line is still on track to be released sometime in the April - June timeframe."
IF it where to be released in April then there should have been more reviews (sorry, previews).
IF it comes out in May then there is no need to do a preview this early (except to gain marketing hype).
IF it comes out in June then a "preview" this early should have contained technical information about the chip with a "review" containing benchmarks at a later date closer to its release date.
jojo4u - Thursday, April 7, 2005 - link
Was Cool'n'Quiet enabled at the power draw test?fitten - Thursday, April 7, 2005 - link
#81, Intel publicly stated that implementing HT on the P4 was about a 5% increase in logic/area on the chip. While it may be "ugly" and such, 5% certainly isn't that much of area when you consider how much area other functionality takes up.#82, I haven't seen anyone announce duel core products. Dual core products, however, should be out very soon from Intel (and are [p]reviewed in the article this discussion is attached) for example). ;)
Kagjes - Thursday, April 7, 2005 - link
I must say i really can't relate to any of these multitasking scenorios. Well, except the teamspeak-gaming part.What would be most interesting for me is dual-core, dual monitor setup. That would ROCK, and i'm really suprised noone remembered that. For instance, playing a divx on 1 monitor while gaming on other. Or watching TV on 1 while gaming on other (i do that a lot, if the football match is boring). These tasks are pretty much repeatable and can be measured. I was thinking about getting another monitor for these kind of purposes, but i'm not sure on how would my comp react to that kind of stress.
justly - Thursday, April 7, 2005 - link
My personal thoughts about publishing an article on a duel core that is not yet availabile, nor will be for awhile, is that it is strictly marketing hype similar to what happened with the 1.13 GHz P3 or any other product that we the consumer have to wait for.The idea that duel core would have a benifit in multitasking or multithreaded apps could have easily been demonstrated on a multiprocessor system using Opterons or Xeons. So now we get a review full of new (wow factor) benchmarks describing the new found benifit of having two processors working together.
I have no real problem with "preview" style articles, but having a plethora of benchmaks in a "preview" is pointless unless the goal is to advertise (hype) a product. The only people I see having any use for this "perview" are the ones that desperately want proof that their manufacturer of choice (this time being Intel) is superior, or those people who are so fickle that they will support the newest thing on the block no matter if it is needed, affordable or even the best option at the time it becomes available.
Quanticles - Thursday, April 7, 2005 - link
Ive heard that HT is very ugly to impliment, you have to tear everything up. I'm not sure HT is very useful when you have dual core anyway.fitten - Thursday, April 7, 2005 - link
#47, if HT is simply a "bandaid", then why is AMD the only major CPU vendor not using it? IBM uses it heavily in their Power5, Sun is making their next CPUs (Niagra) very highly SMT (same thing as HT). Arguably, both of those architectures have much more shallow pipelines than the P4, yet see reason to provide SMT. AMD is the only holdout.dougSF30 - Thursday, April 7, 2005 - link
Make that 2.4GHz in Q3 (not Q4) for the DC Opti 2.4GHz.dougSF30 - Thursday, April 7, 2005 - link
DC A64 samples are at 2.4GHz, not 2.2GHz.DC Opteron is supposedly launching at up to 2.2GHz in a couple weeks, and will be available at 2.4GHz in Q4.
DC Opteron fits in 95W TDP, DC A64 gets a 110W TDP.
Both are apparently 2 x 1MB L2, per the Italian site that put the CeBIT-timed article about the DC A64 sample.
dparish - Thursday, April 7, 2005 - link
Great article. I am confused by one piece:From page 8:
Seagate Barracuda 7200.7 NCQ - 25.2 minutes
Seagate Barracuda 7200.7 no NCQ - 33.6 minutes
Western Digital Raptor WD740 - 30.9 minutes
I'm shocked that the WD740 is slower. Isn't this the NCQ enabled 10,000 rpm drive? I would have expected this to beat out the 7200.7 NCQ seagate. Any reason why the WD740 would be slower?
Another point / comment:
Is it possible that DVD Shrink is slower on the AMD's because it has not been compiled / optimized for the AMD? Would this be any different on the 64 bit version of XP if DVD Shrink was compiled for 64 bit? Of course this may be moot as DVD Shrink is not longer being developed by the author.
-Dave
Umbra55 - Thursday, April 7, 2005 - link
Why do you still use the nForce 4 reference board for AMD tests? The real nForce 4 boards exist since over a quarter now and perform 10% better than the reference board. You also pretend to compare CPUs with similar price. Wouldn't it make more sense to compare combinations of mobo/CPU/memory of similar price?This is not a fair comparison (Or is it the intention?)
Umbra55 - Thursday, April 7, 2005 - link
yde - Thursday, April 7, 2005 - link
In this atricle, I was unable to find cache information about the AMD chip (usually equipped with a 512 KB L2 cache).To my point of view, in multiple threads scenario, the cache size may have dramatic influence and may explain several handicaps in the benchmarks. It would be nice to know what happens with a 1MB L2 cache Athlon 64 to keep things equal.
AMD chip has shorter branch prediction lines and seem quite well equipped for multitasking in theory, so why is it appearing so weak?
snorre - Thursday, April 7, 2005 - link
We're still waiting for proper benchmarks comparing dual core Smithfield with dual Opteron/Xeon. When will we see this?Comparing dual core CPUs with single core CPUs is like comparing apples and oranges, totally meaningless.
Calin - Thursday, April 7, 2005 - link
Thanks for the minimum frame rate comparison! And maybe you should not use so many flash-heavy pages, especially considering that you already told us that Athlon64 is much slower in Flash than Pentium4...Looks like Pentium D is a better choice in more ways than the Prescott is
xsilver - Thursday, April 7, 2005 - link
The multitasking gaming analysis combined with HT and NCQ was very insightful, almost pioneering...and with dual core AMD's people have to remember that the whole architecture is designed differently, and hence the possible suprises in tasks that the pentium D may not perform well on
but the power consumption advantage on amd is mighty tempting as amd will be the cpu that just keeps on saving -- with power bills that is
Azsen - Thursday, April 7, 2005 - link
Hi Anand,I was thinking, seeing most games are single-threaded you might like to try and benchmark this scenario on the dual core machine:
Set all the processes including operating system processes and other background processes to run on CPU 0 (first core). Then set the particular game to run on CPU 1 (second core). Have nothing else but the game running on CPU 1. This should dedicate a whole CPU core to the game for maximum performance in theory. I believe you can set the affinity in the task manager or use a batch file to do it.
Then run a Doom3 benchmark or HL2 benchmarks to see if the gaming performance is increased by letting the single-threaded game have a whole CPU core with no interuptions.
Would this be feasible to test?
Cheers. :)
AnnoyedGrunt - Wednesday, April 6, 2005 - link
The AMD not competing with the PD 2.8 will be due to price.Remember that AMD will need to sell a 2 core 2.2 GHz processor for about double the price of the single core version (maybe even more if yields are a problem). Therefore, you should expect a dual core 2.2 proc to be closer to $500-600 instead of the ~$300 processors tested here.
This also explains why AMD is focusing on the workstation market first. It's an area where price is typically not as much a factor, and where they are competing with Xeon prices, so it will be much easier for them to sell procs @ the higher prices.
I'm guessing that AMD's desktop dual core procs will start @ 1.6 or 1.8 GHz, and be priced on par with the Intel offerings. I think the overall performance will be similar to the current single core behavior, but now you can have more stuff running.
I currently have an AMD 939 3200+, so I am looking forward to their dual core offerings in a couple years, and hoping my mobo will be compatible. That would be very cool.
Also, I would like to add vote for a WoW test with Teamspeak running in the background as well as an istance if IE or Firefox with tthotbot.
Thanks,
D'oh!
SLIM - Wednesday, April 6, 2005 - link
Hi Anand,Any chance you could give us a little peak at what dual opterons on the nforce pro chipset can do in the same benches that you did in this article.
A configuration something like this:
Opteron 252s clocked at 1.8GHz, 2.2GHz, 2.6GHz
1 GB ECC DDR400
NVIDIA nForce pro motherboard
ATI Radeon X850 XT PCI Express
NCQ enabled HDD (maxline III, 7200.8, etc)
I looked through all the old articles I could find, but most seemed to contain only server oriented benches. I think that kind of article would be very enlightening as to the future performance of dual core athlons (probably within a couple percent)... maybe help put to rest some of the questions a lot of us have about waiting for amd dualcore, intel dualcore or just overclocking the heck out of a venice a64.
SLIM
Jep4444 - Wednesday, April 6, 2005 - link
SSE3 does almost nothing for the Venice which is barely faster than the Winchester(Xbit Labs benchmarked it already)While Anandtech said the dual cored A64s wouldn't compete with the Pentium D in encoding, unless they actually have dual cored A64s which they can't show us, i'd be willing to argue with that. Encoding is one of those things thats largely affected by Hyper-Threading and the Pentium D loses hyper threading. When the A64 goes dual, its performance increase will be larger than Intels.
GregL - Wednesday, April 6, 2005 - link
P.S. I love your site... been reading it for years now.GregL - Wednesday, April 6, 2005 - link
Anand,Thanks for the explanation and the quick reply.
Have an excellent day,
Greg
Anand Lal Shimpi - Wednesday, April 6, 2005 - link
GregLLast time I checked (which admittedly was a while ago), SMP support was broken in the later builds of Q3A. I can't remember if it was Quake 3 or the combination of Q3 and ATI/NV drivers, but the performance stopped improving.
Take care,
Anand
GregL - Wednesday, April 6, 2005 - link
I know Quake3 is dated but how about a quick benchmark with the new dual core CPU. Quake 3 is supposed to support dual core.seta r_smp "1"
Thanks,
Greg
Goi - Wednesday, April 6, 2005 - link
I didn't know 50 cent was an avid reader of ATTuborg - Wednesday, April 6, 2005 - link
It`s nice to finaly see some competition from Intel.They slapped together theyr old stuff in a new package. But we all know that a new package isen`t going to change anything(Like wrapping s*** in gold paper).
Be happy as longe as it last, and have your 15min of fame.
Remember they rushed out the dual core, and they did it for you IndelJugen!.
Viditor - Wednesday, April 6, 2005 - link
Thanks for the article Anand...none of us take Charlie seriously anyway..."AMD's dual core will be quite impressive, even more so than Intel's"
I am hearing the same. There is some serious research work being done in the TV and Film industry right now with the dualcore Opterons, and it is MOST impressive! Still under NDA (as are we all), I can only say that the results so far have been much better than expected!
Son of a N00b - Wednesday, April 6, 2005 - link
Thank You Anand for the great article, especially the info on the NCQ. Great writing, and overall a very good read.btw, I understand how fusterated you must feel making these benchmarks, not having things work, trying to remember all the things you want/have to do next, ect......Keep it up Anand, that is why you are the best!...try to get some sleep though m8 :-P
I would probably say that the 2.2 ghz from AMD it would not be compared to the dualCPU in this article because if the 2.2 is going to probably be the FX line, then it would be compared with the top of the line of Intel's...remember this was an article about "value" dual cores (oxymoron ;-))...so due to price and probably performace it would not be paired with the Pentium D at 2.8...sort of like AMD's naming scheme, an AMD 2800 at 1.8 ghz matches up with a 2.8 ghz Intel...so I would assume that AMD's biggest baddest dualy will blow Intel out of the water...and not because i am an incessant AMD fanboy because i am an avid gamer, but becuase of AMD's past performace, and AMD architechure is designed for dual core. We shall see, wh shall see...
*STATEMENT: The author of this post is not hereby responsible for any grammatical errors, typing, or syntax, of any kind.* lol
Googer - Wednesday, April 6, 2005 - link
In adition to my #57 post,In the future I cannot Imagine the power requrements of CPU's they may end up needing their own 500watt dedicated supply and a second one for HDD's, GPU's, Fans, motherboards, and accessories.
Googer - Wednesday, April 6, 2005 - link
The only thing more ineffieciant than a 250watt fully loaded Prescott is the old eniac, It was said that when it was turned on the Whole City of Philadelphia would go in to a brown out. I am afraid that modern processors are taking steps back instead of forward.stephenbrooks - Wednesday, April 6, 2005 - link
#55 yes, that would be my best bet on what Anand was implying. AMD's 90nm process is impressive on power efficiency.bdchambers79 - Wednesday, April 6, 2005 - link
Hmm... if permormance isn't the only metric for dual-core AMD, then perhaps power also plays a role? What if they could craft a 2.2gHz monster that had the same power draw as a 2.0gHz a64? Such a beast, besides being a processing power-house, would show extreme overclock potential.cHodAXUK - Wednesday, April 6, 2005 - link
'The move down to 90nm really cut down AMD’s power consumption a lot, to the point where the 90nm Athlon 64 3500+ actually consumes less power under full load than the Pentium 4 630 at idle.'ROFL, I expect the old Prescott space heater gags to start again after this review :)
Crassus - Wednesday, April 6, 2005 - link
Hey Anand,on page 13 top diagramm it says "Thanks to DVD Shrink behaving and running with a low priority, our gameplay was largely unaffected on the Athlon 64. The performance dropped less than 3% in Doom 3. ..." yet the graph shows the A64 at the bottom far behind the Pentiums. Is the graph messed up or the comment?
Great article though and thatnk you for listening to your readers.
Thanks also for the NCQ-page. I still wonder if you want to pick up where you left off some months ago and look into RAID in more detail, possible measuring the impact of NCQ on the benefits of SMT or vice versa and CPU-load issues with simple RAID5 setups on dual-core?
Cheers, Crassus
Lonyo - Wednesday, April 6, 2005 - link
#50 and JeffLet's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz.
AMD's dual core will be quite impressive, even more so than Intel's. Don't look at performance as the only vector to measure though...
....PRICE.
I think you were spot on. 2.2GHz DC Athlon is almost certain NOT to compare to PD2.8GHz.
The article is comparing 3 systems with the same price processor.
The comments from Anand suggest that the 2.2GHz AMD DC will blow the 2.8GHz PD away, but at a higher price.
The 1.8GHz AMD DC is more likely to be comparable (and possible faster, if the 2.2GHz Athlon really is amazing). But who knows?
coldpower27 - Wednesday, April 6, 2005 - link
Tomshardware was able to overclock the Pentium EE 3.2GHZ Dual Core to 3.8GHZ, and 4GHZ on some exotic form of cooling.coldpower27 - Wednesday, April 6, 2005 - link
Rumors indicate AMD's aiming for a launch at 1.8GHZ - 2.2GHZ frequencies for the Athlon 64 Dual Cores, I would guess, that most likely that these Dual Cores are based on the San Diego cores, as they each have access to 1MB of cache with a die surface area slightly larger then Clawhammer.Another guess would be they would use the 2.2GHZ Dual Cores as their FX flagship and have the 1.8GHZ/2.0GHZ variants as their mainstream line.
However with AMD saying 2H 2005 for their Dual Cores, we won't have to wait too long then for Intel's new version of Dual Core in 1H 2006 on the 65nm process.
Maybe the 2.2GHZ Athlon 64 Dual Core is much more expensive then the 2.8GHZ Pentium D? Or it competes with Intel's Pentium D 3.2GHZ.
defter - Wednesday, April 6, 2005 - link
Great review! It is refreshing to see different kinds of benchmarks.The problem with dual core Athlon64 is that it won't be launched until 2006 according to the AMD's roadmap: http://www.amd.com/us-en/Processors/ProductInforma... (dual core Athlon64 FX will be launched in H2 2005, but no regular dual core Athlon64 is scheduled for 2005)
And in 2006, Intel will have 65nm dual core chips available...
PrinceXizor - Wednesday, April 6, 2005 - link
For those saying that Intel will have ANOTHER dual core system out because of the delay by AMD...remember...this was a technology PREVIEW. There is no released hardware yet.Its going to be interesting to see what the benefits of AMD's built from the ground up to be multi-core approach vs. Intel's patch job (I still have utmost respect for Intel, they created one fine "patch job" under a severe time constraint, sometimes it does pay to be a behemoth ;)) will be.
P-X
NetMavrik - Wednesday, April 6, 2005 - link
Don't take this post the wrong way, I am not knocking Intel's HyperThreading, but I think most people have actually forgotten what it is. It is not a feature of the Pentium 4, but a bandaid approach to keep their now 31 stage pipeline busy doing something. The reason that a 3.8GHz P4 doesn't run circles around a 2.6GHz A64 is that horribly designed 31 stage pipeline. AMD can't implement some form of HyperThreading because it doesn't have a bunch of processor stages sitting around doing nothing like a P4 does without HyperThreading.segagenesis - Wednesday, April 6, 2005 - link
#45 - The Athlon 64 still lags behind on encoding cabability and its been shown over the past year. However some of the content tested was designed specifically for P4 optimization so its hard to get a reliable result. Who knows... until we see the new AMD64 core with SSE3 we cant really pass judgement.Quanticles - Wednesday, April 6, 2005 - link
AMD needs to send Anandtech one of their dual-cores with a DFI nForce4 SLI board.That'd shut a lot of people up.
If AMD owns single thread, then why wouldnt they own dual core. People would be throwing money away buying intel's dual core.
Spearhawk - Wednesday, April 6, 2005 - link
The first graph on the DVD Shrink/Game bench still seems a bit suspect. Why would the P4 outperform the PD and the A64 when under normal circumstances the A64 should be superior in singlethread and the PD in multithread?Anway, great article. I'm really looking forward to seeing what AMD has to offer since while I'd greatly like improved multitasking I'd also like a good gaming CPU.
Marlin1975 - Wednesday, April 6, 2005 - link
What about overclocking the Dual core chip???Regs - Wednesday, April 6, 2005 - link
Ah, I mis read it. You used the Intel 955X.I have to ask then, is it because of AMD's onboard memory controller that they don't have to manufacture another motherboard for the dual core CPUs? If so, you think AMD was thinking about this scenero (dual cores) well before the first clawhammer came out?
That would sound impressive. Compared to Intel's dual press hots.
segagenesis - Wednesday, April 6, 2005 - link
#37 - Well im hard pressed to really want one when my current setup is still sufficient. That and I have my heart set on AMD64 for gaming. Even at DVD backup I can do a movie in about an hour even with full quality under DVD Shrink. And really, I would use DVD Rebuilder which is very much single threaded and in my book Quality > Speed. Takes about 6-7 hours but better results than DVD Shrink could have wished for.Anand Lal Shimpi - Wednesday, April 6, 2005 - link
RegsThe nForce4 Intel Edition platform was unrelated to the Intel dual core platform, it was just something I was working on at the time.
The platform arrived DOA, I'm guessing it's an error on NV's part.
Take care,
Anand
Jeff7181 - Wednesday, April 6, 2005 - link
"AMD's dual core will be quite impressive, even more so than Intel's. Don't look at performance as the only vector to measure though... "You like to tease us, don't you? :)
Regs - Wednesday, April 6, 2005 - link
Hey Anand, I noted that you said in Part one that the Intel Dual core was one of the most stable config's you ever had. However in part two and quote, "After recovering from Part I and realizing that my nForce4 Intel Edition platform had died, I was hard at work on Part II of the dual core story. "Was this human error or was it a manufacturing error?
michael2k - Wednesday, April 6, 2005 - link
#26: Actually, encoding a DVD should be a multi-tasking event in of itself!Task1: DVD creation; menu's, transitions, etc
Task2: DVD encoding; background rendering of menu's and transitions
Task3: DVD encoding; background rendering of the actual menu
Task4: Burning of DVD; you should be able to start burning the DVD before the encoding in task 2 or 3 is complete, as long as the burner is properly buffered
segagenesis - Wednesday, April 6, 2005 - link
#30 - Excuse me for trying to save money also. Last I checked Intel was still more expensive. Not to mention Extremely Expensive edition.rqle - Wednesday, April 6, 2005 - link
All i know is that i alt-tab / alt-enter to the desktop running general apps all the time while gaming. I bought two system so i can download while gaming on the other system for this very reason. To do both at the same time would cause the ftp software to go into idle state with the fastest download speed at only 8-10kb/s. I can set the ftp software at a higher priority but then it would just cripple my gaming. These dual core look very promising, but ill hold out for amd dual core.GentleStream - Wednesday, April 6, 2005 - link
I'm interested in benchmarks that would be relevant to scientific computing and software development.How about benchmarking a parallel compile of some non-trivial software package such as building the
gcc compiler. That takes quite a long time on my 4 year old laptop.
danidentity - Wednesday, April 6, 2005 - link
So when are the Pentium D and the new chipsets being released?Spill it Anand. ;)
Turnip - Wednesday, April 6, 2005 - link
#23What about the fourth option? That by the time AMD's desktop dual core processor is available, Intel will have a new dual core processor available. Now, whether we're talking more than "two cores bunged on a chip", or whether we're simply talking a jacked up FSB (which has, remember, always given Intel a hefty jump in the encoding arena), I don't know. But I do know one thing...
Intel is a very big company and Intel has very big sleeves. ;)
Anand Lal Shimpi - Wednesday, April 6, 2005 - link
SpearhawkGood catch, the graphing engine didn't regenerate those graphs properly. Fixed now.
Take care,
Anand
Questar - Wednesday, April 6, 2005 - link
I love this, all the AMD fanboys having seizures over that fact that an Intel CPU can actually have some benefits.It's been a blast reading all these posts the last two days.
Spearhawk - Wednesday, April 6, 2005 - link
He said that a 2.2 GHz dualcore Athlon 64 wouldn't compete with the 2.8 GHz Pentium D at encoding. Notice the encoding part, he said nothing about other stuff.I'm guessing one can know that by looking at dual CPU Opteron systems, the dualcore A64 won't beat them, and if they can't beat a 2.8 Pentium D then the dualcore A64 won't be able to either.
Is there something wrong with the graphs in the DVD Shrink/Game test? The comments doesn't seem to match them (especialy the part about the minimum frame rate being equal)
PorBleemo - Wednesday, April 6, 2005 - link
How do you calculate the system wattages like that? I have been attempting to find detailed information on how to do this but have turned up nothing yet.Thanks!
Illissius - Wednesday, April 6, 2005 - link
#10 - you are quite correct. anyone who games with a processor-intensive background task running at the same time _on a single core processor_ is insane. the reason I wanted to see benchmarks is to see whether dual core changes that.theoretically, I don't see why it wouldn't work:
- you only have one GPU, and only the game is using it
- you have two processor intensive tasks - the game and the background task, and two cores, one for each
hence, no conflict. whether that actually holds up in the real world is/was the question (if the background task is multithreaded, or heavily uses reasources other than just the processor, then naturally the above doesn't hold true).
segagenesis - Wednesday, April 6, 2005 - link
Since I mostly play games ill stick to buying the AMD64 3500+. Thanks. My definition of multi-tasking is using a whole other computer ;)The Pentium D seems pretty decent at multitasking as you would define running two things at once but I rarely do that sort of thing since its kind of dumb to encode a DVD in the background while playing a game. Or does encoding a DVD really interfere with browsing the web? I dont know... that and the heat factory output as if it was bad enough is now worse.
Woodchuck2000 - Wednesday, April 6, 2005 - link
#23 - I'm assuming that a dual core A64 at 2.2GHz will blow a Pentium D out of the water at any of the launch frequencies! The Prescott core isn't really designed for multi-core operation, and needs some kind of arbitration logic and some funky-memory-controlling-thingy to work. As a result, the performance improvements in multi-threaded applications aren't anything like the theoretical extra 100% another core could bring. With A64 being designed for multi-core operation I'd expect the increase in performance to be nearer 85%.As regards the performance gap between the P4 630 and A64 3500+, the majority of the benchmarks shown here are designed specifically to show performance improvements in multi-core processors. The 630 is hyper-threaded and therefore logically multi-cored, if not physically so. As such the 630 will have artificially high performance compared with the 3500+ - in most single-application benchmarks, the AMD chip would thrash both Intel chips.
Is there any chance of adding benches for the 630 with HT disabled (or at least giving us an idea of performance.)? We've got a vanilla A64 versus a HT 630 and Dual-core system. It'd be good to see how a single core performs for reference.
Anand Lal Shimpi - Wednesday, April 6, 2005 - link
Jeff7181I really can't say more, but you are barking up the wrong tree with those assumptions :)
AMD's dual core will be quite impressive, even more so than Intel's. Don't look at performance as the only vector to measure though...
marcusgarcia
We did look at HT performance when it came out, but the problem is that HT doesn't improve performance in all cases. Look at the Gaming Multitasking Scenario 2 tests, HT reduced performance significantly - most likely because DVD Shrink and Doom 3/Splinter Cell were both contending for floating point resources that were in use. Dual core solves this problem by having two complete sets of execution units, so there's no worry about contention between threads for shared resources.
As far as Half Life 2 goes, it is still single threaded so its performance characteristics would be no different than what you see here.
mlittl3
I've been looking into running VoIP or some sort of voice chat program in the background, the problem surfaces in trying to put together a reliable, repeatable workload. Dual core will most definitely help there, but how much - I do not know.
I haven't given up yet :)
BruceDickenson
Glad to have you on-board and thanks for the kind words :)
Woodchuck2000
The new dual core chips are still LGA-775, but they do require a new motherboard (unlike AMD's solution which just requires a new BIOS). Currently Intel's 945 and 955 chipsets will support dual core, and tomorrow I should have a nForce4 SLI Intel Edition board that will support dual core as well. The new NVIDIA chipset does support dual core, but it's up to motherboard makers to implement support for it in their designs.
Check with the motherboard maker to make sure that dual core is explicitly supported by the board, it should say so somewhere in the manual or on the box.
Take care,
Anand
Jeff7181 - Wednesday, April 6, 2005 - link
#22... Anand said a 2.2 GHz Athlon 64 won't compete with the 2.8 GHz Pentium D. That either means a 2.2 GHz dual core Athlon 64 will have lackluster performance, or it will be AMD's new enthusiast line like the FX is right now, which means it would be competing with the Extreme Edition chips, not the regular line.I guess there's a 3rd possibility. He was referring to dual core Opterons which obviously won't compete with the Pentium D any more than the Opteron competes with the Pentium 4 right now.
Woodchuck2000 - Wednesday, April 6, 2005 - link
Just out of interest, does anyone know what socket do the new cores use? Will the new nVidia chipset support the new cores (it was hinted at briefly, but not stated explicitly...)?#19 - What's your source for those assertions? I've heard reports that AMD have got samples running at well over 2GHz and since the K8 architecture is natrually better suited to multiple cores I'd have expected blistering performance. BTW, does anyone know if the AMD cores will be based on the new Venice rev? Is SSE3 a given?
BruceDickenson - Wednesday, April 6, 2005 - link
Hey all, long time reader, this is my first "post"/comment...Just had to say this is one of the most interesting articles I've read in a long time. I loved the NCQ tangent, it almost felt like you were part of a conversation when you read how Lal Shimpi analyzed the anomaly in his testing.
Loved it! Thanks AT!
marcusgarcia - Wednesday, April 6, 2005 - link
14# again.Forgot to say in the last post, my rant is about HT, not dual cores.
I know 2-cores won't make THAT difference on these trivial things (who needs another 2.8ghz for simple stuff?)...but..HT is benefiting GREATLY from it, yet noone mentioned it and didn't even try this sort of test when HT was launched.
When you see the 3.0 HT doing better than a AMD 3500+ (supposedly 500 points faster), you gotta ask how badly would it beat the AMD 64 3000+, which happens to cost almost the same than the P4 3.0 ghz...which happened to destroy the much faster AMD on the test.
That pretty much sucks and leave us with the impression that people either:
a - wanted to benefit AMD
or
b - were too ingenuous to think on these tests when doing HT tests (which can't be true because i always wanted them)
Jeff7181 - Wednesday, April 6, 2005 - link
Errr... correction...1.) The dual core 2.2 GHz Athlon-64's will be less than impressive and won't even perform in the same class as a Pentium D @ 2.8 GHz.
2.) The mainstream Athlon-64 dual core chips will be running at much less than 2.2 GHz, and the 2.2 GHz dual cores will be the FX line, which compete with the Extreme Edition Pentiums.
marcusgarcia - Wednesday, April 6, 2005 - link
14#Completely wrong.
1º: Outlook checks 8 pop accounts for mail and apply it's rules to it every minute or so.
2º: MSN with webcam can eat quite some CPU, specially because i play on the dark with the "low light filter" turned on, which happens to eat quite some CPU.
3º: For every file opened/closed both the AVG and the MS anti-spyware are going to have a check if that's malicious and if the action is allowed.
When i close everything and run 3dmark01 i get around 300 - 600 more points out of it from my 12200 points score.
PS: don't forget IE, which is usually opened here or on tom's hardware (or both and some more), which happen to have a lot of those huge flash banners.
I think that DOES make *a lot* of difference.
Add to that the fact of many people using skype while gaming, mainly on FPS and RTS, which can make all the difference.
Jeff7181 - Wednesday, April 6, 2005 - link
"Let's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz."So you leave two possibilities.
1.) The dual core 2.2 GHz Athlon-64's will be less than impressive and won't even perform in the same class as a Pentium D @ 2.8 GHz, but rather dual core Extreme Edition chips.
2.) The desktop dual core Athlon-64's will be running at much less than 2.2 GHz.
mlittl3 - Wednesday, April 6, 2005 - link
Hey Anand,A really cool multitasking scenario for gaming would be running a game with something like Skype in the background. Everyone saying that a respectable gamer (whatever that means) would not run multi applications in the background is not thinking about VoIP.
I am in Louisiana and I like to game with my friend in Georgia. We talk to each other using Skype will playing Halo on the same server. I know the overhead necessary for VoIP must slow things down some.
Won't dual core help in this case?
Aikouka - Wednesday, April 6, 2005 - link
I know I personally have a lot of things open when gaming, especially if I'm playing World of Warcraft. I'll typically alt-tab out of the game to check IRC or Firefox (with a bunch of tabs open) to look something up or if I'm bored, just browse the net a little bit.The only problem I ever have with slowdowns is if the game is highly CPU-bound and uses up 100% of my CPU, which WoW does almost all the time.
Rand - Wednesday, April 6, 2005 - link
#10- I'm inclined to agree, but people did request it so presumably some people ar interested in doing so for whatever reason."I don´t close AVG and MS Antispyware and MSN and outlook and IExplorer everytime i open warcraft or half life 2, so...WHO MADE ME BELIEVE AMD WAS FASTER?"
Merely running applications in the background isn't going to do much to benefit DualCore/SMP unless those applications are actually utilizing the processor. Odds are MSN/Outlook/Spyware/Anti-virus probably aren't doing a thing but sitting idle when your gaming.
marcusgarcia - Wednesday, April 6, 2005 - link
#10: There surely is.I play warcraft III online, which is a RTS game.
Being so, not all actions are dependant on my reflex (in fact i can many times minimize the game for around 20 seconds which is the time my char takes to walk to a certain place on a given map).
That being said, i am ALWAYS with a few instances of internet explorer open, MSN open, outlook express open and of course AVG and MS Anti spyware loaded on memory with real time protection.
Add the fact that sometimes i am viewing and being viewed on MSN webcam.
I'm sure MANY more people do that.
Remember not all players are FPS gamers...in fact, FPS is far behind MMORPG in sales, which doesn't require near as much attention and reflex.
marcusgarcia - Wednesday, April 6, 2005 - link
OK.Something is very wrong here.
I mean, WHY DIDN'T ALL SITES DO THESE TESTS WHEN HT WAS LAUNCHED?
It clearly shows here what is MUCH better when it comes under regular usage.
A Pentium 4 3.0 ghz is beating AMD's trash on 3500+....i mean, WTF?
Almost noone (does anyone at all?) goes closing all applications before gaming or doing any other activity and HT is clearly giving AMD a serious beating on the multi-tasking scenario (read: EVERYONE's usage).
I don´t close AVG and MS Antispyware and MSN and outlook and IExplorer everytime i open warcraft or half life 2, so...WHO MADE ME BELIEVE AMD WAS FASTER?
I mean...dude...are we talking about servers here to compare single threaded performance?
Are we still on Windows 3.11?
By the way, how in the hell aren't they including Half Life's 2 performance?
Surely the physics engine plays quite a bit on processing and even more surely it is done on separate threads, which would show the dual core being strong even on a single application, let alone on a multi-tasking one.
I'm quite repented for having an Athlon 64 3000+ as my CPU right now when the Pentium 4 3.0 HT would be clearly outspacing the Athlon in every respect as long as i was multi-tasking/opening/closing/minimizing things (e.g.: ALWAYS).
Damn at all these sites.
boban10 - Wednesday, April 6, 2005 - link
i think that this real-world multitasking testing done by Anandtech is 1000 times better than one syntetic benchmark, that is most time optimized for one or another cpu....someone agree ?
ronaldo
mbhame - Wednesday, April 6, 2005 - link
I'm sorry but I find the premise of Page 11 borderline absurd. I *cannot* fathom there's a respectable amount of gamers that actually do that on a regular basis.Anand Lal Shimpi - Wednesday, April 6, 2005 - link
WooDaddyMulti-core multitasking is already quite difficult, you have no idea how frustrating last weekend was. The issue is that I can sit with you on a computer and show you all the areas that dual core will improve performance, but quantifying it so I can stick a bunch of bars in a graph is far more difficult. AMD and Intel are actively working with BAPCo on SYSMark 2006 that should be much more multi-core friendly, but until then we're left with a lot of hard work. We're trying to write our own benchmarks as well, it's just that they take quite a bit of time to put together.
Take care,
Anand
Anand Lal Shimpi - Wednesday, April 6, 2005 - link
Thanks for pointing out the graph error, the labels just got messed up it looks like; should be fixed now.Remember AMD is talking about a 2H 2005 launch for dual core Athlon 64 on the desktop, don't expect to see reviews of desktop parts anytime soon.
As far as the encoding comment goes, it's tough for me to actually elaborate without stepping into areas I can't get into just yet. Let's just say that the dual core Athlon 64 running at 2.2GHz won't be compared to a dual core Pentium D running at 2.8GHz.
Take care,
Anand
WooDaddy - Wednesday, April 6, 2005 - link
Ok Anand, either you're slick or you're slipping. History shows you're slick..You said a dual-core A64 won't help in encoding apps. I know you're not one to say stuff just because you THINK it's true, but because you KNOW so. I'm not at T0M H4rdware..
So.. Since you're alluding to it, WHEN'S THE DUAL CORE A64 TEST COMING OUT!?!?! *pant**pant*
Seriously though, I see that this multicore, multi-tasking benchmarking is going to get quite difficult. How do you know just how fast it really is considering all the combinations of different apps you will have running in the background? It Madonion or those other benchmarking guys going to be coming out with a synthetic benchmarking tool to gauge the max performance of these new multi-core processors?
JoKeRr - Wednesday, April 6, 2005 - link
the power consumption is a series mofo here.hm, with a P4 EE or PDEE, with 2 6800Ultra or GT, you're definitely expecting an oven case.
Wonder how those dualcores overclock-->that should help to make up the single thread performance. On the other hand, Clearly HyperThreading is GOOD, but AMD still says NO.
Jeff7181 - Wednesday, April 6, 2005 - link
Great article. Well done guys.The only thing I have to question is how a dual core Athlon-64 at encoding. On the last page you say, "For encoding performance, you still can’t beat the Pentium D. Even a dual core Athlon 64 isn’t going to help enough in that area."
What makes you think that when the 2.2 GHz Athlon-64 is RIGHT behind the 3.0 GHz Pentium 4 630?
Everything I've read so far has told me that the Athlon-64 should scale better than the Pentium 4 as far as dual cores are concerned... so... what exactly are you basing your opinion that even a dual core Athlon-64 won't help in encoding???
Lonyo - Wednesday, April 6, 2005 - link
Does http://www.anandtech.com/cpuchipsets/showdoc.aspx?... page have an error?It lists the PD 3.2, P4 3.73 EE and the PD 3.2 EE, not the 3 CPU's montioned in the article. Wrong graph maybe.
Rys - Wednesday, April 6, 2005 - link
H isn't even near P on the keyboard...michaelpatrick33 - Wednesday, April 6, 2005 - link
Holy crispy power and cook my eggs in true multitasking while I surf the net Batmansmn198 - Wednesday, April 6, 2005 - link
I guess it makes sense that NCQ would help when multitasking. I assume that this would be the same on single cores as well? The new focus on responsiveness is a good move IMO. The time to switch between apps and redraw the screen and clicking on menus and buttons is what frustrates me when I am multitasking.