Difference Between Xeon versus Desktop?

[Nebulus]

I saw this article posted by a user in another forum.

http://www.custompc.co.uk/labs/149382/intel-xeon-x3200-series.html

Excerpt: However, the Xeon X3200-series CPUs aren't just Core 2 Quads with a different name engraved in the top of the heatspreader. As Xeons are aimed at servers and workstations, the four prefetchers are tuned for these sorts of applications, rather than standard desktop applications and games. As a result of this subtle tuning, the X3220 was 3 per cent slower than the otherwise identical Core 2 Quad Q6600 in our Media Benchmarks, and 25 per cent slower in games. However, in our Folding@home test and Cinebench R10, the X3220 was 3 per cent faster than the Q6600. Unless you're building a server or workstation, you're better off buying a Q6600 over an X3220.


25% slower in games?... anyone know more about this topic?

I have an X3350 (Q9450 equivalent) on order but now I'm not so sure.

[zanzabar]

i dont think that that benchmark is correct but the x3350 should be the same but some1 should test it

[RunawayPrisoner]

http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/MaxBench.jpg

17198 3DMark06 score... and you're still afraid of the gaming performance of this chip? Yeah... right. :p:

[RedHOT]

Where is the sites actual benchmark showing the 25% difference?

[Nebulus]

http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/MaxBench.jpg

17198 3DMark06 score... and you're still afraid of the gaming performance of this chip? Yeah... right. :p:

The question is, would a Q9450 at those exact same settings get the same score? Perhaps someone on this forum with a Q9450 could test this for us.

[Crankybugga]

You are comparing findings with 65nm counterparts to the difference between 45nm chips?

[RunawayPrisoner]

I have seen Singaporean reviews of the Q9450, and most of them could only hit 480MHz and bench at 475MHz max or something. I'm running mine 475MHz 24/7 now, though, and there's the difference. :) Here's the source:

http://forums.hardwarezone.com.sg/showpost.php?p=28945358&postcount=137

And to think I benched mine at 490MHz FSB... wow... :p:

Edit: There's the 1500 points in the difference between 3.80GHz and 3.90GHz as proof for you, too. If anything, I think these Xeons would clock better and fare lower voltages than the actual Q9450. Sure, there might be a difference in CPU score (different boards, maybe?), but the question is... can they run at 3.80GHz or even 3.60GHz stable 24/7 when their max stable FSB is at 475MHz on an Asus Striker II? And versus a budget G35 board? :)

Edit 2: Just realized... Vista versus XP. 600 points in CPU difference. Now I'm confident that there would be none at all. :p: Heck... I dare them to be able to clock at 490MHz and boot into Vista stable enough to achieve a full run of 3DMark06. To me, it seems like Vista just requires a hell of a lot more voltage (and effort) to even boot... whereas you can sail smoothly into XP anytime. I'm going to try installing 3DMark06 onto XP and see what happens. :)

[zanzabar]

u guys need to use the same ram settings they make a huge difference in the 3d mark score

[Nebulus]

I have seen Singaporean reviews of the Q9450, and most of them could only hit 480MHz and bench at 475MHz max or something. I'm running mine 475MHz 24/7 now, though, and there's the difference. :) Here's the source:

http://forums.hardwarezone.com.sg/showpost.php?p=28945358&postcount=137

And to think I benched mine at 490MHz FSB... wow... :p:

Edit: There's the 1500 points in the difference between 3.80GHz and 3.90GHz as proof for you, too. If anything, I think these Xeons would clock better and fare lower voltages than the actual Q9450. Sure, there might be a difference in CPU score (different boards, maybe?), but the question is... can they run at 3.80GHz or even 3.60GHz stable 24/7 when their max stable FSB is at 475MHz on an Asus Striker II? And versus a budget G35 board? :)

I'm not really questioning the overclockability of each nor their voltages and temps. If you ran your x3350 at 400 x 8 and benched some games and then swapped out your cpu for a Q9450 at 400 x 8 and benched some games, would the results be the same? That is my question. Given that every single condition and piece of hardware would be the same, I would think they would score the same, but the article is saying otherwise. If course the article is comparing the 65 nm Q6600 with its equivalent 65 nm xeon. Still, its a question worth investigating don't you guys think?

[zanzabar]

some people said that about the opteron too but it was just bad testing that got most of those scores

[Nebulus]

some people said that about the opteron too but it was just bad testing that got most of those scores

Very true. For good testing, the ONLY variable should be the cpu: x3350 or Q9450. So to truly confirm/disprove the claim in the article, we need someone with BOTH chips to test them at identical settings in the same hardware.

[RunawayPrisoner]

Okay... you know what? I think the Xeon is even faster than the Q9450 at the same clock... or at least they perform on the same level. Check this out:

http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/CPUscore.jpg

And compare that to the CPU score of this guy:

http://forums.hardwarezone.com.sg/showpost.php?p=28945358&postcount=137

X3350: 6286
Q9450: 6238

Both at 3.80GHz (475 x 8), both under Windows XP. I think we can rest the case now. :) And I now think I can break 18000 points in 3DMark06 under Windows XP as well... eheheheh...

Edit: and in case you want to keep it up, I think it's time I point out that... the said article doesn't have anything to back its own statement up aside from some statistics. That, and 3% of a difference is not that much IMO.

[Metroid]

RunawayPrisoner@ You know 3DMark is a synthetic benchmark. So we should not take that as proof to retaliate what another user said it. We need more benchmarks as I am convicted what he is saying about the prefetcher's are true, about the gaming, I am still not really sure as I have never seen a game running on a server processor to date.

Metroid.

[RunawayPrisoner]

RunawayPrisoner@ You know 3DMark is a synthetic benchmark. So we should not take that as proof to retaliate what another user said it. We need more benchmarks as I am convicted what he is saying about the prefetcher's are true, about the gaming, I am still not really sure as I have never seen a game running on a server processor to date.

Metroid.

Then here you go. A real game benchmark... and probably the hottest one in the world right now, too:

DX9 all High
http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/CrysisBenchDX988gts.jpg

DX10 all Very High
http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/CrysisBench88gts.jpg

I'd do more games if I have the time to. But sincerely, this thing handles Crysis well enough.

[Nebulus]

Prisoner, the crysis benchmark doesn't tell us anything because we don't have an equivalently clocked Q9450 in an identical system to compare it to.

[RunawayPrisoner]

I compared my results with another user with a E8400 @ 4.25GHz and the favors are to these chips. What I'm trying to say is not that there is no difference or anything but that these chips do their jobs really well... so why should you care about whether or not they have the name Q9450? I owned a E3110 and now I have an E8400, and IMHO, running in the same setup, the E8400 requires more volts to stay stable, and perform somewhat a bit inferior to the E3110 in... every way. What you're worrying about is some myth about the difference (very subtle) between 65nm chips and their xeon counterparts. And I did try to say that the said article did not even have any benchmark to back their own statements up, didn't I? ;) You ordered the chip, so right now, just compare it to the people who own the same chip instead of worrying about something like the Q9450. I'm sure you made the right choice.

Edit: And oh well, I guess I'll tell you this: BIOS 0405 on my motherboard was supposed to support Q9450, yet it did not recognize the Xeon chip. I had to update to 0503 so the board can recognize the chip correctly. So my conclusion is that THERE IS indeed a difference between these two chips, and that may explain why I don't ram into that 480MHz FSB wall like most others. But then... there is still no sign of this chip being outperformed by its Quad counterpart.

[Metroid]

Then here you go. A real game benchmark... and probably the hottest one in the world right now, too:

I'd do more games if I have the time to. But sincerely, this thing handles Crysis well enough.

I like your efforts. I really do but when I said more benchmarks. I was saying made by them(The Professionals in this area). :)

You could do the tests if you had the necessary hardware or if you were a hardware fanatic. I would really appreciate it.

It is a powerful hardware for games. Why do not you use it for our crunching team :p:

Metroid.

[RunawayPrisoner]

I'm only providing future owners of this chip all the information I can provide... so they know what to expect from the chip. :) Of course, I'm quite concerned about this "myth" as well, and that's why I'd like to try and see... although I doubt anyone will buy both this and a Q9450 chip at the same time.

Edit: Oh, and I'm quoting what another user asked before:

Where is the sites actual benchmark showing the 25% difference?

Yeah... I'd like to know that as well. I'm a guy of actual pictures and numbers, not of statistical talks.

And by the way, since we are going with a "myth" from the 65nm quads, let's go back and look at the 65nm quads, yeah?

http://forums.anandtech.com/messageview.aspx?catid=28&threadid=2089621&enterthread=y

After reading that, here's what I gather: Xeon might require less voltage than their desktop counterparts to run stable at the same clock speed, thus making them potentially better overclockers. Also Xeon can withstand higher thermal than their desktop counterparts. This has actually been proven in the case of E3110 versus E8400 processors, so I'd say that this is probably true.

And... back to google.

[Metroid]

I'm only providing future owners of this chip all the information I can provide... so they know what to expect from the chip. :) Of course, I'm quite concerned about this "myth" as well, and that's why I'd like to try and see... although I doubt anyone will buy both this and a Q9450 chip at the same time.

Edit: Oh, and I'm quoting what another user asked before:



Yeah... I'd like to know that as well. I'm a guy of actual pictures and numbers, not of statistical talks.

And by the way, since we are going with a "myth" from the 65nm quads, let's go back and look at the 65nm quads, yeah?

http://forums.anandtech.com/messageview.aspx?catid=28&threadid=2089621&enterthread=y

After reading that, here's what I gather: Xeon might require less voltage than their desktop counterparts to run stable at the same clock speed, thus making them potentially better overclockers. Also Xeon can withstand higher thermal than their desktop counterparts. This has actually been proven in the case of E3110 versus E8400 processors, so I'd say that this is probably true.

And... back to google.



Yes you are right. He did not have any real benchmarks to back up his claiming, also yes that is right as it is a server processor is requires less voltage and it has to be much more efficient once it has to contain heat as it will work 24/7. So what I have yet to make my mind is, it gives much more stability, less voltage, better thermal compound than desktops right? It must have some hidden weakness processing things not related.

Servers processors always have been very good for reliability and much more expensive too. The things may be changing on the right direction, not sure though.

Metroid.

[RunawayPrisoner]

Okay... quad-core server chip versus dual-core desktop chip article here:

http://fanboyreview.blogspot.com/2007/04/article-little-intel-xeon-x3210-that.html

Hmm... not sure how legit that is... but oh well. At least it shows Doom 3 running... a lot better on the Xeon chip. Now... would someone jump in here and tell me that the Doom 3 engine was built to take advantage of quad-core processors?

[Metroid]

Okay... quad-core server chip versus dual-core desktop chip article here:

http://fanboyreview.blogspot.com/2007/04/article-little-intel-xeon-x3210-that.html

Hmm... not sure how legit that is... but oh well. At least it shows Doom 3 running... a lot better on the Xeon chip. Now... would someone jump in here and tell me that the Doom 3 engine was built to take advantage of quad-core processors?

When Doom 3 was launched did not have any quad core support and no support was ever being announced.

Edit: That guy was drunk, see the doom 3 resolution that is why was so much faster.

Metroid.

[RunawayPrisoner]

So much faster at the same clock speed? I'd run the game at the lowest resolution possible, too, if I want to test my CPU out.

Okay... found a professional review after some tedious searches:

http://www.benchmark.co.yu/modules.php?name=BenchmarkNews&file=article&sid=12432&oliver=0

From what I can see, E6600 @ 2.4GHz only outperforms X3220 @ 2.4GHz by a little tiny bit in gaming... almost invisible differences. E6600 was also faster in some cases, too... and I'm just guessing that Q6600 would yield the same result... somewhat. But I don't think it can be up to 25% in gaming. As far as I know, Crysis is the only game that might benefit just a little from quad-core processors.

[Metroid]

So much faster at the same clock speed? I'd run the game at the lowest resolution possible, too, if I want to test my CPU out.

Okay... found a professional review after some tedious searches:

http://www.benchmark.co.yu/modules.php?name=BenchmarkNews&file=article&sid=12432&oliver=0

From what I can see, E6600 @ 2.4GHz only outperforms X3220 @ 2.4GHz by a little tiny bit in gaming... almost invisible differences. E6600 was also faster in some cases, too... and I'm just guessing that Q6600 would yield the same result... somewhat. But I don't think it can be up to 25% in gaming. As far as I know, Crysis is the only game that might benefit just a little from quad-core processors.

I think that guy was drunk, should have had similar results maybe Xeon had something that could slice a bit of performance.

25% is just too much. So we could say both were drunk the guy who did that Doom 3 test and the guy said 25% for desktops counterparts.

We need more benchs. We will have soon. I know it.

Metroid.

[Nebulus]

So much faster at the same clock speed? I'd run the game at the lowest resolution possible, too, if I want to test my CPU out.

Okay... found a professional review after some tedious searches:

http://www.benchmark.co.yu/modules.php?name=BenchmarkNews&file=article&sid=12432&oliver=0

From what I can see, E6600 @ 2.4GHz only outperforms X3220 @ 2.4GHz by a little tiny bit in gaming... almost invisible differences. E6600 was also faster in some cases, too... and I'm just guessing that Q6600 would yield the same result... somewhat. But I don't think it can be up to 25% in gaming. As far as I know, Crysis is the only game that might benefit just a little from quad-core processors.

Well, thats good news; I would go as far as to say that the differences in the gaming benchmarks were well within the margins of error:
968.3 vs 969.7 fps for Quake III... basically the same here
206.6 vs 209.6 fps for Doom III... basically the same (slight edge toE6600)
158.18 vs 161.85 for Farcry... basically the same (slight edge toE6600)

I agree 25% seems a bit far fetched.


We need more benchs. We will have soon. I know it.

Hehe, I think so too. Hopefully when people see this thread tomorrow we will get some more up to date benching done.

[RunawayPrisoner]

Yeah... and I wonder what would be reliable as a benchmark... :p

[zanzabar]

i think that source would be the best since its cpu heavy

[Nebulus]

Figured I would give this thread a bump and see if anyone else has experience/knowledge/ or can officially comment on this supposed difference between xeon gaming vs desktop gaming.

[RunawayPrisoner]

Hmm... one question: supposedly... a E8400 should be faster or on par with a Q9450 in gaming at the same clock speed, right? Unless the game supports quad-core processors. In which case, I'd like to ask if... Crysis supports quad-core processors?

[Nebulus]

Hmm... one question: supposedly... a E8400 should be faster or on par with a Q9450 in gaming at the same clock speed, right? Unless the game supports quad-core processors. In which case, I'd like to ask if... Crysis supports quad-core processors?

At the same clock speed as an E8400, I would expect the Q9450 to be faster, simply because it has a 12 mb cache versus 6mb. As for crysis and quad core support, I'm not too sure. I've heard rumors that is is quad friendly but I can't verify that.

[RunawayPrisoner]

At the same clock speed as an E8400, I would expect the Q9450 to be faster, simply because it has a 12 mb cache versus 6mb. As for crysis and quad core support, I'm not too sure. I've heard rumors that is is quad friendly but I can't verify that.

Well, it's definitely faster (just by a little bit) on my X3350 versus on the E8400. Another member and I had a little competition about that earlier, and I ended up using both my E8400 and X3350 on the same setup to test it out. Incidentally, the X3350 was faster by 1fps although on the E8400 at the same clocks, the graphics card was clocked a bit higher. If the rumor proves true... then at the same clock speed, we'd see the Q9450 outdoes the E8400 greatly in Crysis, maybe by up to 4fps or so.

[Nebulus]

Well, it's definitely faster (just by a little bit) on my X3350 versus on the E8400. Another member and I had a little competition about that earlier, and I ended up using both my E8400 and X3350 on the same setup to test it out. Incidentally, the X3350 was faster by 1fps although on the E8400 at the same clocks, the graphics card was clocked a bit higher. If the rumor proves true... then at the same clock speed, we'd see the Q9450 outdoes the E8400 greatly in Crysis, maybe by up to 4fps or so.

Scratch what I said earlier about cache advantage, it's wrong. If crysis only supports two cores, then a Q9450 and E8400 should perform EXACTLY the same when at equivalent clock speeds. Why? Because I just did some research and the 12 mb cache of the Q9450 is not shared amongst all 4 processors. The Quads are just 2 "core-2-duo" processors on one chip right? So each pair of cores has access to it's own 6mb cache independent of the other.

So if crysis is only dual core supported, those two cores being utilized by Crysis only have 6mb of cache to work with... just like the E8400.

[IanB]

Scratch what I said earlier about cache advantage, it's wrong. If crysis only supports two cores, then a Q9450 and E8400 should perform EXACTLY the same when at equivalent clock speeds. Why? Because I just did some research and the 12 mb cache of the Q9450 is not shared amongst all 4 processors. The Quads are just 2 "core-2-duo" processors on one chip right? So each pair of cores has access to it's own 6mb cache independent of the other.

So if crysis is only dual core supported, those two cores being utilized by Crysis only have 6mb of cache to work with... just like the E8400.

Not necessarily... :p:

Think a bit more. If the E8400 has 6MB of cache total and two cores only, then yes the max amount of cache available to it is 6MB. But how much cache is available to the quad depends on WHICH two cores the game uses. If it is two "adjacent" cores on the same die then the performance and cache use will be identical. But if the cores are on different dies then potentially they can access 12MB of cache between them as in your first suggestion, so there should be a cache boost to performance.

IIRC the OS schedules threads to cores based on core load, so if you can set app affinities so that one core is always less used, and an "opposite" core is the preferred one for the game, then you might be able to force usage of opposite cores like that. Or maybe you can set multiple affinities for a single app, not sure as I don't have a multicore proc right now. :(

[RunawayPrisoner]

That's what I thought as well. And yes, the cache on the X3350/Q9450 chips are 2 x 6MB instead of 4 x 3MB.

But I don't think Crysis performs the same on Q9450 and E8400. Why? Because here are some results of E8400 and X3350 on the same system, with the same setup, and... same kind of benchmark:

E8400:
http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/CrysisBenchE8400.jpg

X3350:
http://i235.photobucket.com/albums/ee60/cuongvietpham/PC/CrysisBench88gts.jpg

Disregarding the higher VGA mem clock with the E8400, you can see a good 1fps speed boost on the X3350 side on run 1. Both CPUs were clocked the same (8 x 475) and were even given the same vCore. So why the difference? And if Q9450 should indeed perform on the same level as E8400, then that means X3350 is... a tad faster than Q9450? We're talking about 1/30 of a difference here, about 3.33% if you ask me. Not to say anything, but at least for now, I am confident that this chip can perform as good as the E8400, and at least it can in Crysis. I also checked this with another member in another thread (as said before), and the results also favored the X3350, but maybe it was because my VGA was faster than his in this game... 9600GT versus HD 3870.

Not necessarily... :p:

Think a bit more. If the E8400 has 6MB of cache total and two cores only, then yes the max amount of cache available to it is 6MB. But how much cache is available to the quad depends on WHICH two cores the game uses. If it is two "adjacent" cores on the same die then the performance and cache use will be identical. But if the cores are on different dies then potentially they can access 12MB of cache between them as in your first suggestion, so there should be a cache boost to performance.

IIRC the OS schedules threads to cores based on core load, so if you can set app affinities so that one core is always less used, and an "opposite" core is the preferred one for the game, then you might be able to force usage of opposite cores like that. Or maybe you can set multiple affinities for a single app, not sure as I don't have a multicore proc right now. :(

Good theory, though. I'll see about trying it out tomorrow... after I've passed Orthos 10K for 8h or so. Some people demand stress testing results... :p: And yeah, I guess I have to affiliate core 0 and core 3 to Crysis. I mean... what are the chances of them being paired up? :lol: And if it doesn't work out, I'll try all other possible pairs (which is 2) before I draw any further conclusion.

[Emerica]

I really don't believe that the Xeon is 25% slower in games, that just doesn't make any sense.

[KTE]

The Xeons are the best chips off the wafer in each bin, most would clock better, at lower temps, power and volts than the LGA-775 counterparts given the same platform, cooling and BIOS. They have far better stability and lower degradation over lifetime too and yup, they are specifically tuned for different tasks native to the server world. Since Intel doesn't make much public mention on this through documentation means we can't really know how and its full effects until rigorously and astutely benchmarked following a fixed methodology, OR, if a devoted review site carries this out for us. I doubt highly there is such a difference between the two as shown in Post#1 with low load applications such as desktop ones; it would matter and show up in server applications properly, esp. TPC-C, Linpack and the like, wherever you have very high core loads with high memory usage scenarios, the Xeon will show more performance optimizations. Maybe they added some latency in the cache algorithms to optimize for aggressive large array prefetching, that's all I'm thinking has been done.

I've used both sets of Penryn at two of my work places running daily but not "benchmarked" the two at all.

As for Crysis CPU dependency, it's a shambles. I've no idea what is wrong with its code but there really is something wrong with its execution multi-thread parallelism and procesor power scaling. Far too erratic. See under benchmarking for instance, Q6600 3.8G should beat 3.2G, but it didn't, and the Phenom 2.6G which is slower at equal clockspeed overall, was equaling the 450FSB 3.6G Q6600 perf. in it - kinda shows how lame the bench is.

And that's by removing any GPU bottleneck, playing 800x600 & medium GFX settings.