Sign in with
Sign up | Sign in
Your question

CPU price/performance graph

Last response: in CPUs
Share
February 20, 2007 9:36:21 PM

Hello forumers,

Here are my own CPU price/performance graphs. Now, obviously I don't have the resources to do testing on these products myself, so I am relying on information gathered from the web. In this respect, I certainly bow to the THG independent analysis. I will try to be open about the techniques I'm using and I am certainly willing to take suggestions.

Okay, so, data: The performance index is the normalised to the X2 3800+ chip, which I believe is an appropriate baseline peformer for 2007. The "general" peformance index is a composite of benchmark scores in:

3DMark06 CPU
LAME audio encoding
Quake 4

The "gaming" performance index is an composite of:

Quake 4 HQ
FarCry
F.E.A.R.
Oblivion 1600x1200

Generally, I've taken the numbers from www.hardware.info. Data has been spliced in from other sources using interpolation. All data can be scrutinised by downloading the .zip file at the end of this post and viewing the spreadsheet.

For pricing, I've used NewEgg. Where a processor is not available at newegg, the price is obtained at TigerDirect or PriceGrabber.

The graphs: I've put price on the x-axis and performance on the y-axis. I've changed the orientation of the graph relative to THG's because this way seems more natural to me. It shows the diminishing returns curve. Points above the curve are good, points below are poor, relative to the family.


General performance


General performance, zoomed view


Gaming performance



Note: The price of the FX-7x series includes two processors and a motherboard correction. The motherboard correction is equal to the price of the QFX motherboard minus the cost of a motherboard capable of running any of the Intel chips. The reference motherboard is, for now, the Gigabyte GA-965P-S3. The price added to the FX-7x series is just over $250 as of 22 Feb 2007.

Download data and graphing MATLAB script here. (updated 2007.04.10)
February 20, 2007 9:52:01 PM

uh...
ok.
February 20, 2007 10:07:01 PM

Sacreligeous!

... if I were to do a graph... i would label the points on the graph... as... say the processor names.
Related resources
February 20, 2007 10:11:20 PM

Ah, yes, I'm working on that. My past experience graphing has been with Excel and MATLAB, and no I'm using OpenOffice, so that's why data labels have evaded me so far. Right now I'm working at adding in some other CPU benchmarks to the mix.
February 20, 2007 10:25:58 PM

i might even suggest putting an opacity on the line of trajectory... perhaps 50% so that we can decern which values lie beneith it
February 20, 2007 10:29:48 PM

Nice effort, but for me, only p/p graphs for *overclocked* CPUs mean anything. After all, if I'm definitely going to OC, why worry about stock performance? Useful graph anyway.
Cheers
Synergy6
February 20, 2007 10:43:49 PM

I always felt conflicted about overclocked numbers. On one hand, it's really the best way to get value for the dollar. I really think that the sweet spot for price/performance has got to be the E4300 (esp. after the upcoming price drop). But I would definitely constrain the overclocked figures to be those with stock cooling, since the added expense of aftermarket CPU coolers isn't included in the price part of the equation.

I'm willing to graph so long as I find the data.

@ Trinitron64: I've added 50% transparency to the trendlines. I agree it looks better now. I'll upload a new graph soon.
February 20, 2007 10:53:18 PM

Thanks for this...

Though, why don't you just use newegg.com to get prices? Thats where everyone buys from practically. I just dislike pricegrabber I guess...

Better than THG that never gets updated but maybe once a month.
February 20, 2007 11:45:12 PM

Yeah, I can do that... it'd be less work anyway.

>> graph's been updated.
February 21, 2007 1:55:30 AM

By using more than one store (pricegrabber), you have introduced a random element into the pricing, and thus into the curves (unless you did your own kind data smoothing like throwing out special sales and price blips somehow).

The ways that occur to me to fix this are to either get all the prices from a great site (big volume, good prices) like Newegg only; OR to use an average from 5 good stores (the same set of 5 stores every time), throwing out the low and high price and averaging the other 3 for instance, though this would probably take more than 30 minutes to do.
February 21, 2007 2:09:29 AM

As requested by two people now, I'm switching my pricing preference to go with newegg. In the case that newegg does not carry a processor at the time I check (such as FX-60 today), then I will fall back on pricegrabber. From the changes I've found, most parts were very close to the same price at both newegg and pricegrabber.

The new prices will be reflected in the next graph update.
February 21, 2007 2:18:06 AM

Unless the pricegrabber price is typical to some major seller, like TigerDirect, etc., you might need to throw it out.
February 21, 2007 2:45:43 AM

Just a warning, Tiger Direct has a terrible rebate policy. I'm still waiting for the rebate on a Celeron D I ordered two years ago. (-_-)*
February 21, 2007 3:40:18 AM

Perhaps just get that chip as an average like I suggested, and the rest from NewEgg.

My last quibble would be to expand the lower and crowded half of the chart.

But, in any case, thanks for the work.
February 21, 2007 4:16:50 AM

New graphs out with newegg prices and I've added back the E6300. Adding this processor involved taking data from other sites. Since the benchmarks aren't exactly the same (such as the file encoded by Lame), I've had to do a bit of points mapping. For example, I would find a ratio between site X's score for 3 processors already on my list, establish that there is consistency, and then take the average. Then I would multiply site X's score for the E6300 by this ratio to get the number I've used. This will undoubtedly add a degree of error to the values assigned, but I expect that this error is on the same order of magnitude as the precision of the tests (repeatability).

@ halbhh: I hope that isn't your last quibble. Your feedback has helped me evolve my technique. Thank you for your input.

Cheers
February 21, 2007 7:46:23 AM

Quote:
I'm using OpenOffice, so that's why data labels have evaded me so far.


How in the freakin' hell do you do a search and replace for hidden characters in Open Office? In Word you just search for ^t (tab) and replace with ^p (paragraph break). I must have read 72,394 OO tutorial sites and still can't figure it out!!! :( 
February 21, 2007 12:48:22 PM

Looks good, please keep updating it... like every time a price changes or a new proc comes out... this way we can look at this instead of an outdated THG chart.
February 21, 2007 1:28:30 PM

Quote:
How in the freakin' hell do you do a search and replace for hidden characters in Open Office? In Word you just search for ^t (tab) and replace with ^p (paragraph break). I must have read 72,394 OO tutorial sites and still can't figure it out!!! :( 


Not to turn this into an OpenOffice discussion, but to answer your question, I don't know. I've never searched for hidden characters even in Word. My experience with OpenOffice writer so far is better than their spreadsheet. Excel is very slick. I look forward to the next release of OpenOffice and I hope they keep adding features. I suppose if I was a good programmer I could contribute myself.

@ Buckiller: I'll try to keep the graphs updated. As one other poster noted, he'd like to see some overclocked results. I'm keeping my eyes peeled and I already have half the data I need for overclocks on the E6300 and 3800+ chips.

It seems to me that AMD has an over-abundance of chips available for <$300, and this really crowds the lower portion of the graph. I've added a zoom-view of this section, but I haven't even included a lot of AMD's offering yet. I think I will try to establish a strategy, like showing mostly the 512KB cache chips and maybe just one of the 1MB chips at a given frequency so the added value of double cache can be seen.

I'd like some feedback on the benchmarks that I've used. Two synthetics, one encoding and one game does not seem as well-rounded as it possibly should be. It would be best for me, though, if the benchmarks could come from a site that has done a significant portion of the chips so that I can apply the interpolation-mapping scheme to the remainder.
February 21, 2007 5:00:21 PM

HotFoot you should tell BM that AMD has too many cheap processors.
February 21, 2007 5:24:49 PM

"No, I don't think we'll be telling them THAT!"

Graphs updated. They're now generated in MATLAB. I've added the 5600+ and 6000+ processors.
___________________________

I've now added two overclocked processors. My goal is to keep the overclocked processors at stock cooling so to not affect the price. I'm not sure if it's possible, though, to hit 3 GHz with the X2 3800+ on stock. Any opinions?

The overclocked processors are not included in the curve regression. The idea is to demonstrate the added value of overclocking. For instance, the "value added" by overclocking the E6300 is over $200, since it requires at least an E6600 processor to match the performance at stock and that chip is more than $200 more expensive.

I've also dropped the PCMark05 benchmark from the mix, since I found a very consistent ratio between the PCMark05 scores and 3DMark06 for the categories of dual and quad core. 3DMark06 showed better scaling from 2 to 4 cores, and the scores are more widely available online, so this benchmark is kept.
February 22, 2007 1:38:44 PM

I'm impressed with the data! Interesting!

If you have more ambition for more, a 3rd column in the numbers table could be the ratio: performance/price (either this or price/performance, and either will need a clear note like (higher is better or lower is better), and then deciding whether or not to order the chart on this or on the performance or the price.

Another possible thing I'd look at (if you haven't already!) is the appearance of the graphs if you use true zeros at the axis intersection. This would tend to emphasize how close many processors are perhaps, which is relevant actually, but it also helps clarify the performance/price ratio visually. If it tends to collapse the data points onto each other, then I understand.
February 22, 2007 1:55:16 PM

Another quibble. First, note my previous post just above.

I was surprised seeing the 4x4 processors looking to have a *much* better ratio than the EX6800 (being aligned on the chart, but higher performance!), then I realized the prices might not have a normalized motherboard cost association, which actually matters for the 4x4 chips. If an average MB cost is thought to be $110, and the 4x4 MB is $360, then you have to add $250 to the 4x4 price to normalize it to the other processors.
February 22, 2007 2:09:28 PM

This is a general problem for all of the comparisons, and especially the FX-7x series as you've said. I don't know if there's a simple solution.

For one thing, if you're running an AM2 chip at stock, you can get an extremely cheap motherboard. Budget motherboards for C2Ds tend to be a little more expensive. Motherboards capable of overclocking well tend to be around $110 and up. So, while I've tried to keep my OC data restricted to what should be possible on stock cooling in order to avoid needing to include the cost of an aftermarket cooler, the extra cost of using a motherboard that's good for overclocking isn't taken into account. The situation is even worse for the FX-7x series.

What else needs to be considered is the RAM. The AMD processors need the best possible RAM to maximise their scores, while the C2D can achieve most of its potential on cheaper, slower memory. I would consider that the memory and motherboard factors cancel each other out, but it would take a lot of reasoning to try to prove this in any useful way.

Another point is that two of the benchmarks I'm using, 3DMark06 and LAME, scale well with the number of cores, and only one benchmark, Quake, does not. This makes the graph favor quad-core processors. I'm thinking it might be better for me to split the data into gaming/everything else, since people spending $1000 on a processor are often looking for the highest possible FPS, and the fact that the processor handles LAME well is an bonus.

It would be interesting to consider price/performance at a system level, but seriously things can get out of hand.
February 22, 2007 2:25:57 PM

Good points. I tended to agree re the Ram costs vs MB costs re AMD vs C2duo, but lately the DDR2 800 is on sale and about the same as DDR2 667, so the advantage there is now to AMD.

I do think the price/performance ratio in the numbers table would be rather important, since many readers here are willing to state (recently) that the C2duo is better in this regard, when the opposite has been true for 3 weeks on much of the curve (where most of the processors are).

I do think the 4x4 FXs should indeed have about $200-250 added on their price, since this amount is much larger in degree compared to other approximations.

I am still curious as to the look of the graphs with true zeros!

But, all in all, great work! It's a very interesting OP, and should become a sticky, IMO.
February 22, 2007 2:28:19 PM

Quote:
But I would definitely constrain the overclocked figures to be those with stock cooling, since the added expense of aftermarket CPU coolers isn't included in the price part of the equation.

I'm willing to graph so long as I find the data.


I think one way to handle that is to pick a good cooler and track OCing of the various chips that use that cooler.

Now that may be difficult, given the wide variety of coolers out there, but I figure that if you pick a Zalman, Thermalright or Skythe, you're probably set.

The other thing is that coolers can outlast a CPU. For example, I bought a Thermalright XP-90 2 years ago, and when I finally figure out which CPU I'm getting (4300 for $163 or 6400 for $159), I'll move it to the new CPU. It'll cost me $6.00 for 975 adapter, but that's not a big deal. Then again, I bought the cooler at least as much for it ability to quietly cool as it's ability to improve OCs.
February 22, 2007 2:33:39 PM

I have to agree that the extra cost for the dual-socket motherboard should be added to the FX-7x series. I'll get right to this. I've also added a price/performance column to the table, and that will be updated soon. What I'm looking for before the next update is a 3DMark06 CPU score for the X2 5200+ processor at stock settings.
February 22, 2007 3:10:33 PM

The graphs have been updated to reflect the adjustment of the FX-7x price to include a premium for the motherboard. Also, I have added a price/performance column to the table.

And for halbhh, since you want to see what the graph looks like with the origin showing, here you go. I don't know that it's all that informative, but if you're looking at p/p ratio then this does better highlight the value of the cheaper processors.

February 22, 2007 3:10:34 PM

anyone else NOT seeing a graph or a link to a graph?
February 22, 2007 3:16:57 PM

Quote:
anyone else NOT seeing a graph or a link to a graph?


For some strange reason, I was having this problem yesterday. I'm running firefox 2.0, and I looked into the problem, but I couldn't figure out what it could have been. It seems that for a lot of places I was visiting, images from 3rd party sites wouldn't load. Now, in older versions of firefox, you had the option to not load 3rd party images, but this is gone in 2.0.

Today, the images are showing for me, so I no longer have to switch computers to be sure my images are uploading properly. What I did yesterday that must have fixed the issue is turn off and then on again the option Tools>>Options>>Content>>Load images automatically.

I had played around disabling adblock, but that didn't do anything for me. I ended up uninstalling adblock and installing adblock plus anyway. That could have fixed it, but it doesn't make sense to me.

I hope this helps you.
February 22, 2007 4:13:37 PM

There is no need to include the price of aftermarket coolers depending on how much you OC your system.

Example - The Stock Cooler for the C2Duo is designed to handle the 2.93Ghz of the E6800. So long as you OC the E4300 only to this point it is operating fully within design specs.

So you Get a 1.1 Ghz OC and no need for added cooling costs.

Only when you push the CPU/Cooler beyond design specs do the aftermarket coolers add to the cost.

The aftermarket coolers are as much for noice reduction as they are for OCing.
February 22, 2007 5:27:55 PM

Good concept and methedology. More data, more processors, more OCed processors, and make the graphs easier to read. Keep it up. Don't disinclude processors to make the graph more readable, that's silly. We're in a digitial environment here, perhaps release the graphs themselves instead of pictures of them? That "Less than $300" range that is so "crowded" is that way because that's where most people shop: fill it up.

Oh, and don't worry about stock cooling for OCed processors. YOu haven't included them in the trendlines and the trendlines don't include motherboard and ram prices either. It's up to the person building the system to do the reasearch and factor in cooling costs for OCing and re-run the price/performance themselves. Just focus on the CPU, less work for you and more data for us. I'd say avoid benchmarks with exotic cooling, but any reputable benchmark using conventional air cooling is fair game to be included. Great idea to not include the OCed data in the trend lines, if you get enough info you might want to make a third trend line just for OCs.

One other idea: don't make Intel vs. AMD trend lines. Just make one trend line. That way it will be easier to see which processors, regardless of brand, fall above and below the line. One trend line for stock, one for OCed (and just assume that OCed systems are probably using extra cooling reletively cancelling the extra cost out and don't worry about it). The p/p graph isn't about Intel p/p vs. AMD p/p, it's about performance period vs. MY bank account. Anyone else out there want to second this? Could you make a quick graph with only one trend line so we can see what it looks like?

It's also great that you included processors in the list and just put "0" in if you didn't have benchmark data for them. You should put even more processors in that list and encourage people to help you find that data or even run the benchmarks you like themselves and get you the results with screenshots and such.

All the the AMD processors listed are dual-core, yes? It should say so. And you should feel free to include a few single-core processors as well for reference.

In genearll I would say don't worry as much about your methodology being perfect and your graph being accurate as with disclosing what your methodology is and presenting more information. You manifesto stated one of your primary reasons for creating you own database was because THG was too slow to update and didn't provide enough info, don't slow yourself down trying to be perfect or you'll end up in the same place (or doing a LOT more work). The graph will never be perfect and it only takes very limited portion of computer performance into account anyway. Present more information and let the users look at the details so they can make adjustments for their specific situation.

Ok, and one more radical idea: The trend line is very smooth, this is not realistic. The price/performance points on processors jump around, so should the trend line. Would it be difficult to calculate a trend in segments so that it went up and down and visually illustrated any "sweet spots". We can easily see single processors that are sweet by seeing how far they are above the trend, but could you make a line that shows the trend of where the sweet spots are? Eh, perhaps this wouldnt be very useful in practice with such a small range of data points and many of the processors coming in close to the trend, just throwing it out there.

Once long ago THG had a giant CPU chart with large varieties of different chipsets, types of ram, and amounts of ram listed. It wasn't as accurate for quickly determining reletive CPU performance if you held all other factors as equal as you could, but it was a hell of a lot more usefull as it included many more processors and also included data on how chipset and RAM affected benchmark scores and the relative performance between a new system and an older system. When you build a system there is a lot more to it then just a CPU so it's kind of silly to try and pretend otherwise when comparing CPUs. You'll end up putting more effort into creating a less useful (although arguable more accurate) guide if you do.

Oh, in your table of CPUs it would be nice to link each processor to the benchmark(s) you pulled the data from since they may be using different test beds. Kind of like citing your sources since you're drawing info from a variety of sources so we can quickly go and compare the raw data between where you put points on the graph.

Ok, and now for a very specific request to add info for
1. OCed e4300 ~@2.4ghz and ~@3ghz
2. OCed x2 3600+ Brisbane ~@2.4ghz and ~@2.8ghz
(this is what I'm currently debating for myself, not many other people seem to be but I think they should!)
February 22, 2007 7:24:16 PM

Quote:
The graphs have been updated to reflect the adjustment of the FX-7x price to include a premium for the motherboard. Also, I have added a price/performance column to the table.

And for halbhh, since you want to see what the graph looks like with the origin showing, here you go. I don't know that it's all that informative, but if you're looking at p/p ratio then this does better highlight the value of the cheaper processors.



Yeah, I did enjoy seeing it with the zeroes. Thanks!

The *interesting* thing is.... the 4x4 FXs with the expensive motherboard are *still* a better buy than the dual core extreme.

That's kinda interesting.
February 22, 2007 10:55:03 PM

The QFX processors are a better by than the X6800 if you're doing a lot of well-threaded applications. It's not too hard for the four cores to overtake the two. However, in gaming benchmarks, the X6800 beats all of the quad-cores. This is one of the main reasons I think I'll produce a separate "gamers" p/p graph.
February 22, 2007 11:14:44 PM

Quote:
The QFX processors are a better by than the X6800 if you're doing a lot of well-threaded applications. It's not too hard for the four cores to overtake the two. However, in gaming benchmarks, the X6800 beats all of the quad-cores. This is one of the main reasons I think I'll produce a separate "gamers" p/p graph.


Yes, just what we should expect. But....unless the x6800 is more than 15% better on the games than a closely priced FX (accounting the $250 MB delta into it), then the FX would be a better buy for a couple of reasons: we do more than games, and that MB will allow upgrade to an 8-core platform with a new arch soon.

I didn't expect to be pointing out any advantages to the 4x4! But considering the new arch drop-in upgrade barcelona....it's a whole different picture!
February 22, 2007 11:23:29 PM

You've made quite the post there! There are many good suggestions.

Quote:
Good concept and methedology. More data, more processors, more OCed processors, and make the graphs easier to read. Keep it up. Don't disinclude processors to make the graph more readable, that's silly. We're in a digitial environment here, perhaps release the graphs themselves instead of pictures of them? That "Less than $300" range that is so "crowded" is that way because that's where most people shop: fill it up.


I try to add a little information each day. Today I tired, but failed, to find a 3DMark06 score for a X2 5200+ processor. I have the other scores I need for that one. As far as releasing the graphs themselves, I believe I can save the figures in a MATLAB format (*.fig). That should allow you to pan/zoom all you want yourself. However, I don't know where I can upload these for sharing. Right now I'm generating .png images and having those hosted on ImageShack, which is a simple enough system. The same goes for my tables of data. The easiest thing for me to do would be to just share my spreadsheet, but it's a matter of how. I'm sure it's simple, I just don't know yet.

[quote}Oh, and don't worry about stock cooling for OCed processors. YOu haven't included them in the trendlines and the trendlines don't include motherboard and ram prices either. It's up to the person building the system to do the reasearch and factor in cooling costs for OCing and re-run the price/performance themselves. Just focus on the CPU, less work for you and more data for us. I'd say avoid benchmarks with exotic cooling, but any reputable benchmark using conventional air cooling is fair game to be included. Great idea to not include the OCed data in the trend lines, if you get enough info you might want to make a third trend line just for OCs.[/quote]

I guess I can agree with that. People who OC are going to be more advanced users anyway. Any OC on air cooling is now fair game.

Quote:
One other idea: don't make Intel vs. AMD trend lines. Just make one trend line. That way it will be easier to see which processors, regardless of brand, fall above and below the line. One trend line for stock, one for OCed (and just assume that OCed systems are probably using extra cooling reletively cancelling the extra cost out and don't worry about it). The p/p graph isn't about Intel p/p vs. AMD p/p, it's about performance period vs. MY bank account. Anyone else out there want to second this? Could you make a quick graph with only one trend line so we can see what it looks like?


I'm up for this idea. I've always found any discussion of the intersection of the trend lines to be meaningless. Good deals will be self-evident.

Quote:
It's also great that you included processors in the list and just put "0" in if you didn't have benchmark data for them. You should put even more processors in that list and encourage people to help you find that data or even run the benchmarks you like themselves and get you the results with screenshots and such.


Yes please, people find me the information I'm missing. Regarding the benchmarks I like, well, I'm just using a few common benchmarks. No one has yet suggested anything different than what I'm using, but the hardest part (at least at first) is finding the benchmark data. Updating prices later will be easy.

Quote:
All the the AMD processors listed are dual-core, yes? It should say so. And you should feel free to include a few single-core processors as well for reference.


Yes, all the AMD processors are either the X2 or FX series. I cut out the "X2" from the AMD chip names because of the amount of clutter. A graph does still need to be readable. I'm looking at trying to make the labels a lighter colour in the background so that the points will stand out better. I'll just add a note to the OP that the processors are all from the Intel Core 2 (desktop) and AMD X2 and FX lines.

Quote:
Ok, and one more radical idea: The trend line is very smooth, this is not realistic. The price/performance points on processors jump around, so should the trend line. Would it be difficult to calculate a trend in segments so that it went up and down and visually illustrated any "sweet spots". We can easily see single processors that are sweet by seeing how far they are above the trend, but could you make a line that shows the trend of where the sweet spots are? Eh, perhaps this wouldnt be very useful in practice with such a small range of data points and many of the processors coming in close to the trend, just throwing it out there.


It wouldn't take that much work for me to set up what you're describing. Basically I'd do a cubic spline fit on to some tavelling average created from the data. It might not add value to the graph, though. I'll see about giving it a try and posting the result.

Quote:
Oh, in your table of CPUs it would be nice to link each processor to the benchmark(s) you pulled the data from since they may be using different test beds. Kind of like citing your sources since you're drawing info from a variety of sources so we can quickly go and compare the raw data between where you put points on the graph.


If I just make my spreadsheet available you'll be able to find what you're looking for.

Quote:
Ok, and now for a very specific request to add info for
1. OCed e4300 ~@2.4ghz and ~@3ghz
2. OCed x2 3600+ Brisbane ~@2.4ghz and ~@2.8ghz
(this is what I'm currently debating for myself, not many other people seem to be but I think they should!)


I'll look for some data. So far, though, I haven't even found enough benchmarks for the X2 3600+ at stock settings to include it. I will keep looking.
February 22, 2007 11:29:18 PM

Quote:
Yes, just what we should expect. But....unless the x6800 is more than 15% better on the games than a closely priced FX (accounting the $250 MB delta into it), then the FX would be a better buy for a couple of reasons: we do more than games, and that MB will allow upgrade to an 8-core platform with a new arch soon.


The X6800 is 23% and 40% faster than the FX-74 and FX-70, respectively, in the Quake 4 benchmark. The FX-62 is faster than the FX-70 and FX-72 in games, and only slightly behind the FX-74 (which it really ought to be considering it's clocked slower). In fact, the X2 6000+ scores higher in Quake 4 than the FX-74.

More on games later...
February 23, 2007 12:21:55 AM

Since you included the premium for 4x4 mobos, why not include the premium of Intel cpu mobos? They tend to be more expensive.
February 23, 2007 12:22:37 AM

Quote:
More on games later...


And later is now: The OP has been updated with a gaming benchmark graph. The general performance remains the same mix, for now, but the gaming benchmarks are Quake 4 and FarCry (one OpenGL and one DirectX). I hope this better demonstrates what many people are looking for.
February 23, 2007 12:56:27 AM

Quote:
Since you included the premium for 4x4 mobos, why not include the premium of Intel cpu mobos? They tend to be more expensive.


Quote:
This is a general problem for all of the comparisons, and especially the FX-7x series as you've said. I don't know if there's a simple solution.

For one thing, if you're running an AM2 chip at stock, you can get an extremely cheap motherboard. Budget motherboards for C2Ds tend to be a little more expensive. Motherboards capable of overclocking well tend to be around $110 and up. So, while I've tried to keep my OC data restricted to what should be possible on stock cooling in order to avoid needing to include the cost of an aftermarket cooler, the extra cost of using a motherboard that's good for overclocking isn't taken into account. The situation is even worse for the FX-7x series.

What else needs to be considered is the RAM. The AMD processors need the best possible RAM to maximise their scores, while the C2D can achieve most of its potential on cheaper, slower memory. I would consider that the memory and motherboard factors cancel each other out, but it would take a lot of reasoning to try to prove this in any useful way.

Another point is that two of the benchmarks I'm using, 3DMark06 and LAME, scale well with the number of cores, and only one benchmark, Quake, does not. This makes the graph favor quad-core processors. I'm thinking it might be better for me to split the data into gaming/everything else, since people spending $1000 on a processor are often looking for the highest possible FPS, and the fact that the processor handles LAME well is an bonus.

It would be interesting to consider price/performance at a system level, but seriously things can get out of hand.
February 23, 2007 4:19:14 AM

As requested, there is an update including an overclocked e4300. The review is from X-Bit Labs, using a Zalman 9500 cooler. Link.

Also info is now available for the X2 3600+ (stock and overclocked), from X-bit Labs.
February 23, 2007 3:28:05 PM

couple of questions: 3 links work, but the zoom view link does not.

Is it reliable that 4300s can be overclocked so high? If so, you'd think none of the nearby Intel processors should be bought -- just the 4300 would be worth buying.
February 23, 2007 4:34:47 PM

Xbit had to increase the voltage to get to that 3.4Ghz, but they did have a 3Ghz without the increase, which would represent a nice overclock anyway. though I'm not an expert, the overvolting should not be on a computer where it would matter much if the cpu fails. If it's only for gaming and you don't care about risking the $170 (or so/w shipping etc), then why not.
February 23, 2007 6:09:53 PM

Yeah, the zoomed view isn't working.

http://www.filefront.com/ or something similar might be the solution you desire for hosting other files. Or you could set it up for bitorrent or ed2k.

The single trend line looks nice, but it's not what I expected. Did you include the OC data in the trend line?

If that gaming performance graph is at all accurate it's brilliantly eye-opening. Good idea.

Good progress, keep it up. I think I'll try to find some raw data for you a little later today.
February 24, 2007 6:04:26 PM

This OP deserves to be a sticky. Anyone know how to make that happen?
February 24, 2007 7:26:54 PM

You have to PM the mods. But its unlikely.
February 25, 2007 2:28:29 AM

Quote:
http://www.filefront.com/ or something similar might be the solution you desire for hosting other files. Or you could set it up for bitorrent or ed2k.


I've zipped my data/graphing files. If you have MATLAB, you can run grapher.m or just open the *.fig files. Download here.

Quote:
The single trend line looks nice, but it's not what I expected. Did you include the OC data in the trend line?


The trend line is combined for non-overclocked Intel and AMD processors, not overclocked.

Quote:
If that gaming performance graph is at all accurate it's brilliantly eye-opening. Good idea.


I pretty much had to make a gaming graph when I saw how disparate the stats were between general, well-threaded benchmarks and standard games. Hopefully the next batch of games will improve things for the quad-core systems.

Quote:
Good progress, keep it up. I think I'll try to find some raw data for you a little later today.


Thanks for the help! Any luck with the data?[/i]
February 25, 2007 3:45:06 PM

Quote:
You have to PM the mods. But its unlikely.


Why do you think it is unlikely? I did just PM the mod (I only found one mod for this forum).

If the "guide" sticky is more than 3 months out of date, why don't they remove it I wonder?
February 26, 2007 1:56:42 AM

The graphs have been updated. I expect I'll see about updating the prices on either a weekly or a bi-weekly basis. It's really not that much work to get prices anyway. The changes in prices as of this Sunday are few. Two AMD processors got more expensive: the 6000+ and 5600+, by $80 and $114, respectively. The Intel processors didn't move by much. Those that did were down a few dollars.

As for benchmarks, of the stock CPUs I am missing data for the 5400+ (all) and 4200+ (FarCry only). Other than those I have tracked everything down. I am still very happy to accept new data for overclocked processors, as I haven't gotten nearly so far with data on those.

Cheers!
February 26, 2007 4:08:25 AM

Quote:
Since you included the premium for 4x4 mobos, why not include the premium of Intel cpu mobos? They tend to be more expensive.


Quote:
This is a general problem for all of the comparisons, and especially the FX-7x series as you've said. I don't know if there's a simple solution.

For one thing, if you're running an AM2 chip at stock, you can get an extremely cheap motherboard. Budget motherboards for C2Ds tend to be a little more expensive. Motherboards capable of overclocking well tend to be around $110 and up. So, while I've tried to keep my OC data restricted to what should be possible on stock cooling in order to avoid needing to include the cost of an aftermarket cooler, the extra cost of using a motherboard that's good for overclocking isn't taken into account. The situation is even worse for the FX-7x series.

What else needs to be considered is the RAM. The AMD processors need the best possible RAM to maximise their scores, while the C2D can achieve most of its potential on cheaper, slower memory. I would consider that the memory and motherboard factors cancel each other out, but it would take a lot of reasoning to try to prove this in any useful way.

Another point is that two of the benchmarks I'm using, 3DMark06 and LAME, scale well with the number of cores, and only one benchmark, Quake, does not. This makes the graph favor quad-core processors. I'm thinking it might be better for me to split the data into gaming/everything else, since people spending $1000 on a processor are often looking for the highest possible FPS, and the fact that the processor handles LAME well is an bonus.

It would be interesting to consider price/performance at a system level, but seriously things can get out of hand.


I still don't see why you would include the price of 4x4 mobo and not any other mobos?
!