Sorry if its been posted, But the math is wrong :S...
Thats like saying.. you have 3 people (each one represents either memory,core or shader) they are all making wallets, and can each do 10 in an hour.
That means 3 people x 10 wallets per hour = 30 wallets a hour total.
Then you tell them each to work 10% faster...... Therefore..
Wallet maker one does 11 wallets a hour (1.1x10=11)
Wallet maker two also does 11 wallets a hour (1.1x10=11)
And Wallet maker Three also does 11 wallets a hour (1.1x10=11).
This equates to 33 wallets an hour (for a 10% increase from each wallet maker)
This is still overall just a 10% increase in performance.
Now onto the artical, It is impossible to get a performance increase above the amount overclocked... Right?.... The highest overclock here was about 17% on the shader if i remember right
(So by my math even if the core and memory were also 17% overclocks it would still just be 17% faster overall... From my example above)
.. This all makes it definatly impossible to get a 30%+ increase.
Serias you are making the assumption that the core , shader and mem are all performing the same job "Making wallets" and therefore the performance improvement should increase in a linnear fashion
Using your wallet maker example , changing the layout of their workspace by "10%" might lead to a 20% increase in efficiency and corresponding 20% increase in wallets per hour
the shader and core clocks obviously have a multiplicive relationship , not an additive one and the core and shader clocks are obviously helping eachother do their job more efficiently in some manner .
Im sure an Nvidia engineer would be able to answer the "why" of this ;)
Science is based on theory -> experiment -> observation ->conclusion (and maybe another round of experiments if the results dont fit your theory)
In this case your trying to say "but my theory must be right so the observed results are wrong"
Thats like saying "the world is flat so something else must be happening when people circumnavigate the earth , cos it couldnt possibly be that its not flat"
Bottom line you can throw as much theory at this as you like , the
Science is based on theory -> experiment -> observation ->conclusion and if need be - > REVISION
(if the results dont fit the theory , you revise the theory untill it explains the results , THEN retest)
You are correct in how the core, shader and memory are seperate, and how those seperate things are unlinked and affect the framerates in very differnt proportions, I did originally think about this... But I just to make it more understandable... I used a simplified example to show it. It still is incorrect though.
The basic principal is the same, You cannot achieve a larger performance increase the the percentage you originally increased from.
The wallet making example was suitable for simple understanding , Using that example you are correct, If you did for increase workspace by 10% you may get a 20% increase in wallets per hour, the only case where this could happen is where the workspace is the limiting factor, and the increase is because perviously he could only fit say 90% of the needed materials on the desk, whereas now he can fit 100% and work at a much faster rate (seeing as he dosnt need to go back and get more materials half way through.)
But, This does not apply to a GPU efficiency because its just a differnt enviroment.
Im sorry but no matter what, It is definatly impossible to overclock three components by 17% 10% and 10% (or whatever the review said) and achieve a theoretical performance improvement of 50%, with a actual of up to 32%... Perhaps by freak occourance may happen to around 19-20%, But that would be driver issues and inperfections in architecture, No way would 32% be possible!!!
Oh yeah, finally,
Its obviously ridiculous if the clocks have a multiplying increase
(xxx*xxx*xxx = total percentage is not right.)
Let me put this into simpler terms so you can understand....
That is the same as using your analogy...
if you decrease the clocks by 50% on each one you will come out with
0.5*0.5*0.5=0.125
So thats like saying if you run 50% of the speed youll get 12.5% of the performance
Or.. even more stupid...
0.2*0.2*0.2=0.008
If you run 20% of the original core speed you get 0.8% of the original performance
Or the other way round... a 50% clock increase on all components (actually possible on some GPU's) would be
1.5*1.5*1.5 = 3.375
So a 50% increase in clocks means a thoretical increase of 337.5% ..... :S
So can you see now how it is clear that you cannot multiply the percentage increase to determin theoretical MAXIMUM performance... (Maximum reffering to the pervious running at 20% of clock speeds = 0.8% MAXIMUM performance.... (Have to follow same rules for all equations)
If you notice the games also have some customized settings and the 4890 they used was a special for their eyes only manufacturer direct channel non retail version ! LOL
Yes of course, the red rooster BS here is so thick it's PATHETIC.
http://www.anandtech.com/video/showdoc.aspx?i=3555...">http://www.anandtech.com/video/showdoc.aspx?i=3555... ---
"We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts."
---
Yes, caution us way down into page 3 of the 4890 extravaganza review - but just COPY AND PASTE those special card numbers here and NEVER MENTION IT - and put a sappy 703 core on the GTX275 and call it "extravaganza!"
----
Yes, your 275 whomps the 4890 in real life - that's why they have to go through maxxing it out for 4890 here to get what they got - a specially OC'able card from manufacturer - specialized custom game settings and ati favorable games in the majority - a great oc on ati not even achievable retail, and a crappy oc on nvidia that RETAIL BEATS ! http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8... ---
Yeah, just really grewat and valuable data as hell666 says the next page... because he doesn't have a clue, he's bleeding out red rooster all over.
Wow, I guess the guys who programmed WAW and Race Driver did a REALLY crappy job at resource allocation; 30 percent compared to about 8 percent from Left 4 Dead; pretty terrible programming.
I too like the 'change-in-place' resolution graphs, but have to agree that they would be better if the scale was consistent.
As far as the 702mhz OC on your 275, that seems like a weak attempt. The retail evga 275 ftw model for example has been binned as an overclocker and stock speed is 713mhz. My MSI 275 FrozrOC is running at 735mhz right now. I can't seem to find mention of which models of the 275 you were testing with, but obviously not the fastest.
respectfully,...
While I love the 'change-in-place' resolution graphs, they really need to be consistent. Leave games in the same location vertically; and keep the same scale horizontally. That way I can tell at an instant glance what the difference is. I don't like having the range switch from 0-15 to 0-7 to 0-10, plus changing the order of the games, when I click the different resolutions!
After all, the only difference that matters on the graphs is the one the individual bars represent. So why go changing the other aspects? Yes, it's "pretty" to have the longest bar the same length, and to always have the graph sorted longest-on-top; but it makes the graph less readable.
For the few graphs that have a bunch of values clustered near each other, plus one or two outliers, just have the outliers run off the edge. For example, in most of your one-variable graphs, a range of 0-10% would be sufficient. Just make sure that for a given resolution set, the range is the same.
This article completely kicks butt! It includes everything I'd want to see in charts, including both % gains and the actual FPS numbers versus other cards, and all with the three most important resolutions.
Very, very good article. Please keep up this level of quality - the data and the depth really answer all the major questions readers and enthusiasts would have.
Nice job Derek, I've been lobbying for a comparison like this since G80 but nice to see a thorough comparison of the different clock domains and impact on performance.
As I suggested in some of your multi-GPU round-up articles, it'd be nice to see similar using CPU clockspeed scaling with a few different types of CPU, say a single i7, a C2Q 9650 and a PII 955 for example, then test with a fast single GPU and observe performance difference at different clockspeeds.
It'd also be interesting to see some comparisons between different GPUs, say 260 to 275 to 280/285 at the same clockspeeds to measure the impact of actual physical differences between the GPU versions.
It looks like a stock GTX 275 beats the 280 in almost every benchmark even at stock speed. Does anyone have any explanation as to why this is happening??
I guess GTX 280 sales will be dropping quiet a bit now
This whole idea of the three seperate overclocks (core, shader, memory) being able to simultaneously provide almost their full percentage increase to any single result cannot possibly be right.
Imagine you take the situation where a card is overclocked by 10% throughout (instead of 11%, 14%, 18% like you did). Core up 10%. Shaders up 10%. Memory up 10%. Going from your numbers, that would probably have given you about a 20% performance increase in two of the games! Do you really expect us to believe a graphics-card running 10% faster, can give a 20% performance boost to the overall framerate?
How does magically making Core and Shader seperate overclocks allow them to work together to nearly double their effect. If it worked that way, you could split the card up into twenty seperate individually overclockable parts, overclock them all by 10%, and end up with something giving over 3x the performance-- all from a 10% overclock :p
Something else must be happening in addition to what you are doing, and my first priority would be to check the actual speeds the card is running at using a third-party utility which reports not the speed the clocks have been set to, but the actual speed the hardware is running at (I believe RivaTuner does that in real-time in its hardware-monitor charts).
I used rivatuner to check the clock speeds. i made very sure things were running at exactly the speeds I specified. At some clocks, the hardware would sort of "round" to the next available clock speed, but the clocks I chose all actually reflect what is going on in hardware.
I do see what you are saying, but it doesn't work either the way you think it should or the way that you claim my logic would lead it be. Extrapolating the math I used (which I believe I made clear was not a useful judge of what to expect, but an extreme upper bound that is not achievable) is one thing, but that isn't what is actually "happening" and I don't believe I stated that it was.
Like I said, it is impossible for the hardware to achieve the full theoretical benefit from each of its overclocked subsystems as this would imply that performance was fully limited by each subsystem, which it just not possible.
If I was confusing on that point then I do apologize.
Here's what I know, though: 1) the reported clock speeds are the clock speeds the hardware was actually running at and 2) the performance numbers are definitely correct.
I fully realize I didn't do a good job of explaining why the two above points are both true ... mostly because I have no idea why.
I tried to paint the picture that what actually happened was not impossible, while (I thought) making it clear that I don't actually know what causes the observed effect.
Great article. I especially liked the 3 linked graphs. One question though. I've been wondering how much power the lastest graphics cards use when you underclock them to the lowest possible while idling, or does the hardware do it automatically? For example, I have my 2D mode on my 8800gtx set to only 200mhz core/shader/memory using nibitor. Or would it matter?
All the current gen cards do significantly underclock and undervolt themselves in 2D mode. They also turn off parts of the chip not in use.
I believe you can set the clocks lower, but the big deal is the voltage as power is proportional to frequency but proportional to the square of voltage. I don't /think/ it would make that much difference in 2D mode, but then it's been years since I tried doing something like that.
no, you're not getting me. It's listed as a 4980, not a 4890. I thought it was a new card, the "NEXT TIER" if you will. The 900 I thought might be a reference to the number of SP. Once I realized you just transposed the numbers, I got the 900 was MHz.
oooooooooooohhhhhhhhh ... okay ... that's a fun typo. I can't really do search and replace on these graphs either. I'll try and get it cleaned up as soon as I can.
GTX275 and 4890 is really evenly matched in every different way(Price, performance, overclocking performance etc) except for the fact that 4890 can be used with the 19$ Accelero S1:
So what AT has done here is put the 4890 max overclocked in their other article against a low end stock gtx275, then in their extravaganza overclock gtx275 they put up a pathetic overclock that is BEATEN BY An EVGA gtx275 as it arrives !
Worse yet, they jammed their ati 4890 maxxxxx overclocks in for comparison !
------------
How to be a completely biased load of crap by Derek and AT :
max overclock your ati 4890 and put stock gtx275's against it
weak overclock your gtx275 and put maxx overclock 4890's against it
___
ROFLMAO - I HAVE TO LUAGH IT'S SO BLATANT AND PATHETIC.
---
cocka doodle doooo ! cocka doodle dooo ! red rooster central.
Evenly matched? The 4890 OCed beats the GTX275 OCed in almost all benchmarks and wins considerably in every game at the resolution that I play at: 1680x1050. It also uses substantially less power and costs less than $200 (eg. ewiz deal at $160, newegg deal at $180), whereas the GTX275 still costs upwards of $220.
4890 @ 1\1.2 is a really nice overclock. They do mention that the GTX275 did'nt overclock that well.
So I prefer (to be aside of the fanboyism-spectra) to call them evenly matched when talking about performance.
Good for you that 4890 is so cheap over there, here they cost about the same as the GTX275. (280$) :/
Powerdraw from a electrical cost-point of view is unimportant for me, since I have free electricity. (Long live the swedish King! lol..) ;)
But it is better from a heat-point-of-view to have less power-draw of course, yeah, so 4890 is (again:slightly) better than GTX275 at load. Its the other way around for idle though. (I would sincerely call this evenly matched in powerdraw).
I have no clue whatsoever how they compare when it comes to noise, but 4890 is really loud at load, thats for sure. ('But not anymore its not!')
A cursory search on google yieled the highest core overclocks obtainable on gtx 275 boards to be between 700 and 745mhz. If you can show us otherwise, please do.
As for myself, and other hardware enthusiasts I'd imagine, the maximum potential of a card comes from its maximum overclocked performance. The fact that the gtx 275 had a "moderate" maximum overclock compared to the 4890 should not come at the cost of the 4890 in a potential comparison.
I just googled "GTX 275 overclock" and the first article that pops up is from Guru3d which shows the GTX 275 overclocking to 743mhz.
Tweaktown did another one and got 715mhz, but they had no clue what they were doing and left the shaders linked.
Anyway, the point is though, if you're going to do an article on overclocking the GTX 275, why bother with a card that has such poor overclocking capability?
Anandtech's HD 4890 OC article specifically used an HD 4890 that was capable of hitting 1ghz on the core, because not all HD 4890s are capable of attaining such a high core speed.
LOL
You can buy a GTX275 retail at 713 core - and they got theirs all the way up to 703 here ! roflmao
Worse yet they use their 4890 numbers from their specially delivered non retail "golden ATI secret channel" - as Derek the red rooster says here in their 4890 oc extrava article ! - LOL
" We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts. "
http://www.anandtech.com/video/showdoc.aspx?i=3555...">http://www.anandtech.com/video/showdoc.aspx?i=3555... ----
SO THE BIAS IS BLEEDING OUT LIKE A BIG FAT STUCK PIG... IF YOU HAVE ANY SENSE WHATSOEVER THAT IS...
---
a red rooster fanboy like Derek and all his little raging red roosters here love it.
vote me down if you like, but the way this article is worded just reinforces the commonly held assumption that anandtech is biased towards intel/nvidia.
thanks for the feedback ... but I'm not touching overclocked SLI and CF ... ugh!
I didn't include the 4890 1/1.2 in idle power because it is redundant as it doesn't affect idle power. and came in at the same idle power as the other two. I wanted to save on graph space where i could because there was so much data -- plus we already covered that in the 4890 overclocking article. Sorry if I dropped too much out.
Hmm, mkay.
Was just confused by the fact that the sligthly overclocked 4890 wanted less juice than the original version in idle.
Maybe due to better VRM\mosfet underclocking or whatever. :)
At least do GTX275 SLI vs. 4890 CF, (and while doing that, just overclock them slightly, plz ;)
I have my finger on the 'ordering-another-4890-button', but wont buy another until anandtech.com reviews 4890 CF!
" Given the performance and pricing of Phenom II and the upcoming Radeon 5000 series, if AMD does not pull into black [records a profit] and achieves great sales success, we don't know what needs to happen in order for AMD to actually earn some serious money. "
LOL
Will ATI ever make a profit ? We keep hearing how smart they are and how they can really make money while NVidia's monster vore costs nvidia so much ! LOL
Awwww- poor ati can't make dime one while nvidia keeps posting profits...
http://www.brightsideofnews.com/news/2009/6/3/ati-...">http://www.brightsideofnews.com/news/20...-has-the... ----
Any red roosters gonna stop crowing about smaller gpu cores and their savings for ATI who loses money every single quarter, anytime soon ?
Probably NOT - they're so smart, so wise about making gpu's.
Better be careful - from their 4890 extravaganza overclock article they did first and used the results in this comparison against nvidia:
-
" We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts. "
YES WE KNOW - NOW YOU'VE TRANSFERRED YOUR RESULTS HERE AGAINST A 703 CORE THAT IS LOWER THAN THE FTW EVGA GTX275 you can but stock faster!
---
You had BETTER look at a few other places that aren't so GD biased it's built in - and Derek is insanely red rooster fan boyed, and hates Nvidia, obviously, like so many little cluckers here for over a year now.-
---
Jeepers- a special 4890 up against non overclocked gtx275 in the 4890 article, then transfer the special results here, and put NOT a regular clocked 4890 as in the other article to be half fair - but jam the massive results in from the special cards they got...
---
I'm mean you really can't screw it up much worse than that.
They did their best red rooster bias blab possible, I'll say that much for em, and covered it up as best they could.
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.
43 Comments
Back to Article
serrias - Monday, June 8, 2009 - link
Sorry if its been posted, But the math is wrong :S...Thats like saying.. you have 3 people (each one represents either memory,core or shader) they are all making wallets, and can each do 10 in an hour.
That means 3 people x 10 wallets per hour = 30 wallets a hour total.
Then you tell them each to work 10% faster...... Therefore..
Wallet maker one does 11 wallets a hour (1.1x10=11)
Wallet maker two also does 11 wallets a hour (1.1x10=11)
And Wallet maker Three also does 11 wallets a hour (1.1x10=11).
This equates to 33 wallets an hour (for a 10% increase from each wallet maker)
This is still overall just a 10% increase in performance.
Now onto the artical, It is impossible to get a performance increase above the amount overclocked... Right?.... The highest overclock here was about 17% on the shader if i remember right
(So by my math even if the core and memory were also 17% overclocks it would still just be 17% faster overall... From my example above)
.. This all makes it definatly impossible to get a 30%+ increase.
Sorry, But this artical is screwed up.
Tigashark - Monday, June 8, 2009 - link
Serias you are making the assumption that the core , shader and mem are all performing the same job "Making wallets" and therefore the performance improvement should increase in a linnear fashionUsing your wallet maker example , changing the layout of their workspace by "10%" might lead to a 20% increase in efficiency and corresponding 20% increase in wallets per hour
the shader and core clocks obviously have a multiplicive relationship , not an additive one and the core and shader clocks are obviously helping eachother do their job more efficiently in some manner .
Im sure an Nvidia engineer would be able to answer the "why" of this ;)
Science is based on theory -> experiment -> observation ->conclusion (and maybe another round of experiments if the results dont fit your theory)
In this case your trying to say "but my theory must be right so the observed results are wrong"
Thats like saying "the world is flat so something else must be happening when people circumnavigate the earth , cos it couldnt possibly be that its not flat"
Bottom line you can throw as much theory at this as you like , the
Observed results speak for themselves .
Great article btw :)
Tigashark - Monday, June 8, 2009 - link
Meant to sayScience is based on theory -> experiment -> observation ->conclusion and if need be - > REVISION
(if the results dont fit the theory , you revise the theory untill it explains the results , THEN retest)
serrias - Tuesday, June 9, 2009 - link
You are correct in how the core, shader and memory are seperate, and how those seperate things are unlinked and affect the framerates in very differnt proportions, I did originally think about this... But I just to make it more understandable... I used a simplified example to show it. It still is incorrect though.The basic principal is the same, You cannot achieve a larger performance increase the the percentage you originally increased from.
The wallet making example was suitable for simple understanding , Using that example you are correct, If you did for increase workspace by 10% you may get a 20% increase in wallets per hour, the only case where this could happen is where the workspace is the limiting factor, and the increase is because perviously he could only fit say 90% of the needed materials on the desk, whereas now he can fit 100% and work at a much faster rate (seeing as he dosnt need to go back and get more materials half way through.)
But, This does not apply to a GPU efficiency because its just a differnt enviroment.
Im sorry but no matter what, It is definatly impossible to overclock three components by 17% 10% and 10% (or whatever the review said) and achieve a theoretical performance improvement of 50%, with a actual of up to 32%... Perhaps by freak occourance may happen to around 19-20%, But that would be driver issues and inperfections in architecture, No way would 32% be possible!!!
serrias - Tuesday, June 9, 2009 - link
Oh yeah, finally,Its obviously ridiculous if the clocks have a multiplying increase
(xxx*xxx*xxx = total percentage is not right.)
Let me put this into simpler terms so you can understand....
That is the same as using your analogy...
if you decrease the clocks by 50% on each one you will come out with
0.5*0.5*0.5=0.125
So thats like saying if you run 50% of the speed youll get 12.5% of the performance
Or.. even more stupid...
0.2*0.2*0.2=0.008
If you run 20% of the original core speed you get 0.8% of the original performance
Or the other way round... a 50% clock increase on all components (actually possible on some GPU's) would be
1.5*1.5*1.5 = 3.375
So a 50% increase in clocks means a thoretical increase of 337.5% ..... :S
So can you see now how it is clear that you cannot multiply the percentage increase to determin theoretical MAXIMUM performance... (Maximum reffering to the pervious running at 20% of clock speeds = 0.8% MAXIMUM performance.... (Have to follow same rules for all equations)
tenq - Monday, July 26, 2010 - link
China wholesale electronicsChina wholesale from wholesaleonepiece.com
www.wholesaleonepiece.com
mypapayaone - Monday, June 8, 2009 - link
China wholesale electronicsChina wholesale from papayaone
http://www.papayaone.com/">http://www.papayaone.com/
helldrell666 - Saturday, June 6, 2009 - link
Extremely useful Data Mr. Derek.I just love your articles.Your one of the great guys at anandtech who keep anandtech unique in every single way.KhadgarTWN - Friday, June 5, 2009 - link
A great review, but the problem is still there,4890 perform far worse in reality than these canned benchmark
My 4890 has no change on par with GTX 275 at 1920x1200 on most of game.... I have them both on E8500
SiliconDoc - Saturday, June 6, 2009 - link
If you notice the games also have some customized settings and the 4890 they used was a special for their eyes only manufacturer direct channel non retail version ! LOLYes of course, the red rooster BS here is so thick it's PATHETIC.
http://www.anandtech.com/video/showdoc.aspx?i=3555...">http://www.anandtech.com/video/showdoc.aspx?i=3555...
---
"We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts."
---
Yes, caution us way down into page 3 of the 4890 extravaganza review - but just COPY AND PASTE those special card numbers here and NEVER MENTION IT - and put a sappy 703 core on the GTX275 and call it "extravaganza!"
----
Yes, your 275 whomps the 4890 in real life - that's why they have to go through maxxing it out for 4890 here to get what they got - a specially OC'able card from manufacturer - specialized custom game settings and ati favorable games in the majority - a great oc on ati not even achievable retail, and a crappy oc on nvidia that RETAIL BEATS ! http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...
---
Yeah, just really grewat and valuable data as hell666 says the next page... because he doesn't have a clue, he's bleeding out red rooster all over.
Hrel - Friday, June 5, 2009 - link
Wow, I guess the guys who programmed WAW and Race Driver did a REALLY crappy job at resource allocation; 30 percent compared to about 8 percent from Left 4 Dead; pretty terrible programming.MonsterSound - Friday, June 5, 2009 - link
I too like the 'change-in-place' resolution graphs, but have to agree that they would be better if the scale was consistent.As far as the 702mhz OC on your 275, that seems like a weak attempt. The retail evga 275 ftw model for example has been binned as an overclocker and stock speed is 713mhz. My MSI 275 FrozrOC is running at 735mhz right now. I can't seem to find mention of which models of the 275 you were testing with, but obviously not the fastest.
respectfully,...
Anonymous Freak - Thursday, June 4, 2009 - link
While I love the 'change-in-place' resolution graphs, they really need to be consistent. Leave games in the same location vertically; and keep the same scale horizontally. That way I can tell at an instant glance what the difference is. I don't like having the range switch from 0-15 to 0-7 to 0-10, plus changing the order of the games, when I click the different resolutions!After all, the only difference that matters on the graphs is the one the individual bars represent. So why go changing the other aspects? Yes, it's "pretty" to have the longest bar the same length, and to always have the graph sorted longest-on-top; but it makes the graph less readable.
For the few graphs that have a bunch of values clustered near each other, plus one or two outliers, just have the outliers run off the edge. For example, in most of your one-variable graphs, a range of 0-10% would be sufficient. Just make sure that for a given resolution set, the range is the same.
yacoub - Thursday, June 4, 2009 - link
This article completely kicks butt! It includes everything I'd want to see in charts, including both % gains and the actual FPS numbers versus other cards, and all with the three most important resolutions.Very, very good article. Please keep up this level of quality - the data and the depth really answer all the major questions readers and enthusiasts would have.
chizow - Thursday, June 4, 2009 - link
Nice job Derek, I've been lobbying for a comparison like this since G80 but nice to see a thorough comparison of the different clock domains and impact on performance.As I suggested in some of your multi-GPU round-up articles, it'd be nice to see similar using CPU clockspeed scaling with a few different types of CPU, say a single i7, a C2Q 9650 and a PII 955 for example, then test with a fast single GPU and observe performance difference at different clockspeeds.
It'd also be interesting to see some comparisons between different GPUs, say 260 to 275 to 280/285 at the same clockspeeds to measure the impact of actual physical differences between the GPU versions.
spunlex - Thursday, June 4, 2009 - link
It looks like a stock GTX 275 beats the 280 in almost every benchmark even at stock speed. Does anyone have any explanation as to why this is happening??I guess GTX 280 sales will be dropping quiet a bit now
PrinceGaz - Thursday, June 4, 2009 - link
This whole idea of the three seperate overclocks (core, shader, memory) being able to simultaneously provide almost their full percentage increase to any single result cannot possibly be right.Imagine you take the situation where a card is overclocked by 10% throughout (instead of 11%, 14%, 18% like you did). Core up 10%. Shaders up 10%. Memory up 10%. Going from your numbers, that would probably have given you about a 20% performance increase in two of the games! Do you really expect us to believe a graphics-card running 10% faster, can give a 20% performance boost to the overall framerate?
How does magically making Core and Shader seperate overclocks allow them to work together to nearly double their effect. If it worked that way, you could split the card up into twenty seperate individually overclockable parts, overclock them all by 10%, and end up with something giving over 3x the performance-- all from a 10% overclock :p
Something else must be happening in addition to what you are doing, and my first priority would be to check the actual speeds the card is running at using a third-party utility which reports not the speed the clocks have been set to, but the actual speed the hardware is running at (I believe RivaTuner does that in real-time in its hardware-monitor charts).
DerekWilson - Thursday, June 4, 2009 - link
I used rivatuner to check the clock speeds. i made very sure things were running at exactly the speeds I specified. At some clocks, the hardware would sort of "round" to the next available clock speed, but the clocks I chose all actually reflect what is going on in hardware.I do see what you are saying, but it doesn't work either the way you think it should or the way that you claim my logic would lead it be. Extrapolating the math I used (which I believe I made clear was not a useful judge of what to expect, but an extreme upper bound that is not achievable) is one thing, but that isn't what is actually "happening" and I don't believe I stated that it was.
Like I said, it is impossible for the hardware to achieve the full theoretical benefit from each of its overclocked subsystems as this would imply that performance was fully limited by each subsystem, which it just not possible.
If I was confusing on that point then I do apologize.
Here's what I know, though: 1) the reported clock speeds are the clock speeds the hardware was actually running at and 2) the performance numbers are definitely correct.
I fully realize I didn't do a good job of explaining why the two above points are both true ... mostly because I have no idea why.
I tried to paint the picture that what actually happened was not impossible, while (I thought) making it clear that I don't actually know what causes the observed effect.
Kibbles - Thursday, June 4, 2009 - link
Great article. I especially liked the 3 linked graphs. One question though. I've been wondering how much power the lastest graphics cards use when you underclock them to the lowest possible while idling, or does the hardware do it automatically? For example, I have my 2D mode on my 8800gtx set to only 200mhz core/shader/memory using nibitor. Or would it matter?DerekWilson - Thursday, June 4, 2009 - link
All the current gen cards do significantly underclock and undervolt themselves in 2D mode. They also turn off parts of the chip not in use.I believe you can set the clocks lower, but the big deal is the voltage as power is proportional to frequency but proportional to the square of voltage. I don't /think/ it would make that much difference in 2D mode, but then it's been years since I tried doing something like that.
balancedthinking - Thursday, June 4, 2009 - link
Actually, you can save as much as 40W(!)idle when you underclock a 4890/4870. The important part is to underlcock the GDDR5 which automatically adjusts voltage too.http://www.computerbase.de/news/hardware/grafikkar...">http://www.computerbase.de/news/hardwar...ten/ati/...
7Enigma - Friday, June 5, 2009 - link
Anyway you could translate the 4870 portion? Babelfish is not working for me for some reason....jeramhyde - Thursday, June 4, 2009 - link
great article thanks :-) good work with the graphs and setting it out in an easy to follow way :-)my housemate just got a gtx275 today, so we'll be trying those overclocks tonight :-)
nvalhalla - Thursday, June 4, 2009 - link
Alright, I just spent the last 10 minutes looking for a 900 shader 4980 before I realized you meant a 900MHz 4890. It's wrong in every graph.DerekWilson - Thursday, June 4, 2009 - link
It's not wrong, it's just noted in a different way.like I say (702 core) for the GTX 275 with a 702MHz core clock, i say (900 core) for a 4890 with a 900MHz core clock.
I'm sorry for the confusion, but it the graphs were already so crowded that I wanted to save space as much as I could.
nvalhalla - Thursday, June 4, 2009 - link
no, you're not getting me. It's listed as a 4980, not a 4890. I thought it was a new card, the "NEXT TIER" if you will. The 900 I thought might be a reference to the number of SP. Once I realized you just transposed the numbers, I got the 900 was MHz.DerekWilson - Thursday, June 4, 2009 - link
oooooooooooohhhhhhhhh ... okay ... that's a fun typo. I can't really do search and replace on these graphs either. I'll try and get it cleaned up as soon as I can.thanks for clarifying.
walp - Thursday, June 4, 2009 - link
Very nice article as always! :)GTX275 and 4890 is really evenly matched in every different way(Price, performance, overclocking performance etc) except for the fact that 4890 can be used with the 19$ Accelero S1:
http://translate.google.com/translate?prev=hp&...">http://translate.google.com/translate?p...mp;sl=sv...
, which makes it totally quiet and cool. Just watch those VRM-temperatures and you will be just fine!
This is the main reason why I chosed the 4890 over GTX275, and the fact that I had a CF-compatible motherboard.
By the way, why didnt you include the 4890 (1\1.2)GHz idle power draw? Or is it just a little type-o? :)
Request: GTX275 SLI vs 4890 CF (And overclocking bonanza! :)))))
\walp
SiliconDoc - Monday, June 22, 2009 - link
I can't help but use google and check the very first site that comes up:http://www.techspot.com/review/164-radeon-4890-vs-...">http://www.techspot.com/review/164-radeon-4890-vs-...
--
and the GTX275 beats the TAR out of the 4890 !!!
--
I guess derek has once again used some biased bench or the special from the manufacturer 4890, and then downclocked the GTX275 to boot.
WHAT A CROCK !
SiliconDoc - Saturday, June 6, 2009 - link
It's hilarious - the extravaganza overclock they do here for Nvidia can't even match the stock timings of a common EVGA.http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...
-
core 713 derek weakness red rooster nvidia hate 703
So what AT has done here is put the 4890 max overclocked in their other article against a low end stock gtx275, then in their extravaganza overclock gtx275 they put up a pathetic overclock that is BEATEN BY An EVGA gtx275 as it arrives !
Worse yet, they jammed their ati 4890 maxxxxx overclocks in for comparison !
------------
How to be a completely biased load of crap by Derek and AT :
max overclock your ati 4890 and put stock gtx275's against it
weak overclock your gtx275 and put maxx overclock 4890's against it
___
ROFLMAO - I HAVE TO LUAGH IT'S SO BLATANT AND PATHETIC.
---
cocka doodle doooo ! cocka doodle dooo ! red rooster central.
Shadowmage - Thursday, June 4, 2009 - link
Evenly matched? The 4890 OCed beats the GTX275 OCed in almost all benchmarks and wins considerably in every game at the resolution that I play at: 1680x1050. It also uses substantially less power and costs less than $200 (eg. ewiz deal at $160, newegg deal at $180), whereas the GTX275 still costs upwards of $220.walp - Thursday, June 4, 2009 - link
I just wanted to be polite. :4890 @ 1\1.2 is a really nice overclock. They do mention that the GTX275 did'nt overclock that well.
So I prefer (to be aside of the fanboyism-spectra) to call them evenly matched when talking about performance.
Good for you that 4890 is so cheap over there, here they cost about the same as the GTX275. (280$) :/
Powerdraw from a electrical cost-point of view is unimportant for me, since I have free electricity. (Long live the swedish King! lol..) ;)
But it is better from a heat-point-of-view to have less power-draw of course, yeah, so 4890 is (again:slightly) better than GTX275 at load. Its the other way around for idle though. (I would sincerely call this evenly matched in powerdraw).
I have no clue whatsoever how they compare when it comes to noise, but 4890 is really loud at load, thats for sure. ('But not anymore its not!')
\walp
Carfax - Thursday, June 4, 2009 - link
Except that the GTX 275 OC had a very moderate overclock compared to the greater overclock on the HD 4890.I don't see how Anandtech only got 700mhz out of the core.
li3k - Thursday, June 4, 2009 - link
well...A cursory search on google yieled the highest core overclocks obtainable on gtx 275 boards to be between 700 and 745mhz. If you can show us otherwise, please do.
As for myself, and other hardware enthusiasts I'd imagine, the maximum potential of a card comes from its maximum overclocked performance. The fact that the gtx 275 had a "moderate" maximum overclock compared to the 4890 should not come at the cost of the 4890 in a potential comparison.
I stand by my point.
Carfax - Thursday, June 4, 2009 - link
I just googled "GTX 275 overclock" and the first article that pops up is from Guru3d which shows the GTX 275 overclocking to 743mhz.Tweaktown did another one and got 715mhz, but they had no clue what they were doing and left the shaders linked.
Anyway, the point is though, if you're going to do an article on overclocking the GTX 275, why bother with a card that has such poor overclocking capability?
Anandtech's HD 4890 OC article specifically used an HD 4890 that was capable of hitting 1ghz on the core, because not all HD 4890s are capable of attaining such a high core speed.
Why couldn't they do the same for the GTX 275?
This article is B.S..
SiliconDoc - Saturday, June 6, 2009 - link
LOLYou can buy a GTX275 retail at 713 core - and they got theirs all the way up to 703 here ! roflmao
Worse yet they use their 4890 numbers from their specially delivered non retail "golden ATI secret channel" - as Derek the red rooster says here in their 4890 oc extrava article ! - LOL
" We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts. "
http://www.anandtech.com/video/showdoc.aspx?i=3555...">http://www.anandtech.com/video/showdoc.aspx?i=3555...
----
SO THE BIAS IS BLEEDING OUT LIKE A BIG FAT STUCK PIG... IF YOU HAVE ANY SENSE WHATSOEVER THAT IS...
---
a red rooster fanboy like Derek and all his little raging red roosters here love it.
li3k - Thursday, June 4, 2009 - link
I have to agree...vote me down if you like, but the way this article is worded just reinforces the commonly held assumption that anandtech is biased towards intel/nvidia.
DerekWilson - Thursday, June 4, 2009 - link
thanks for the feedback ... but I'm not touching overclocked SLI and CF ... ugh!I didn't include the 4890 1/1.2 in idle power because it is redundant as it doesn't affect idle power. and came in at the same idle power as the other two. I wanted to save on graph space where i could because there was so much data -- plus we already covered that in the 4890 overclocking article. Sorry if I dropped too much out.
walp - Thursday, June 4, 2009 - link
Hmm, mkay.Was just confused by the fact that the sligthly overclocked 4890 wanted less juice than the original version in idle.
Maybe due to better VRM\mosfet underclocking or whatever. :)
At least do GTX275 SLI vs. 4890 CF, (and while doing that, just overclock them slightly, plz ;)
I have my finger on the 'ordering-another-4890-button', but wont buy another until anandtech.com reviews 4890 CF!
\walp
SiliconDoc - Monday, June 22, 2009 - link
Yes the gtx 275 wins even in overclocking... i wonder what went wrong with dereks tests...( no i don't !)...
http://www.techspot.com/review/164-radeon-4890-vs-...">http://www.techspot.com/review/164-radeon-4890-vs-...
SiliconDoc - Monday, June 22, 2009 - link
Not only does the gtx275 beat the 4890, but the gtx260 beats the 4870 !http://www.techspot.com/review/164-radeon-4890-vs-...">http://www.techspot.com/review/164-radeon-4890-vs-...
..
Oh gee, I guess I should follow dereks chart instead... ( GAG ! )
SiliconDoc - Monday, June 8, 2009 - link
" Given the performance and pricing of Phenom II and the upcoming Radeon 5000 series, if AMD does not pull into black [records a profit] and achieves great sales success, we don't know what needs to happen in order for AMD to actually earn some serious money. "LOL
Will ATI ever make a profit ? We keep hearing how smart they are and how they can really make money while NVidia's monster vore costs nvidia so much ! LOL
Awwww- poor ati can't make dime one while nvidia keeps posting profits...
http://www.brightsideofnews.com/news/2009/6/3/ati-...">http://www.brightsideofnews.com/news/20...-has-the...
----
Any red roosters gonna stop crowing about smaller gpu cores and their savings for ATI who loses money every single quarter, anytime soon ?
Probably NOT - they're so smart, so wise about making gpu's.
SiliconDoc - Saturday, June 6, 2009 - link
Better be careful - from their 4890 extravaganza overclock article they did first and used the results in this comparison against nvidia:-
" We absolutely must caution our readers once again that these are not off-the-shelf retail parts. These are parts sent directly to us from manufacturers and could very likely have a higher overclocking potential than retail parts. "
YES WE KNOW - NOW YOU'VE TRANSFERRED YOUR RESULTS HERE AGAINST A 703 CORE THAT IS LOWER THAN THE FTW EVGA GTX275 you can but stock faster!
---
You had BETTER look at a few other places that aren't so GD biased it's built in - and Derek is insanely red rooster fan boyed, and hates Nvidia, obviously, like so many little cluckers here for over a year now.-
---
Jeepers- a special 4890 up against non overclocked gtx275 in the 4890 article, then transfer the special results here, and put NOT a regular clocked 4890 as in the other article to be half fair - but jam the massive results in from the special cards they got...
---
I'm mean you really can't screw it up much worse than that.
They did their best red rooster bias blab possible, I'll say that much for em, and covered it up as best they could.