- Joined
- May 29, 2005
How is ED taking about something off topic? It its a perfectly valid opinion that should be taken as the testing was not done IMO in a property manner as clock speeds being equal do not properly represent each card.
You're*
-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
MSI 7970 vs Evga 680 The real bench results T&B
- Thread starter TURN & BURN
- Start date
How is ED taking about something off topic? It its a perfectly valid opinion that should be taken as the testing was not done IMO in a property manner as clock speeds being equal do not properly represent each card.
You're*
- Joined
- Jun 9, 2010
How is it fair to overclock the 7970 and compare it to a stock 680. Just not very unbiased and rather like a certain end result was intended.
LOL, Frakk... Im not dribbling on about anything, nor am I off subject. I quoted what of yours I responded to. I was trying to help clarify what seems to be a question of yours. Sorry if I misunderstood.
That said, I already conceded, in another thread you were discussing this same subject, that comparing top end overclocks is a valid comparison, but the matching clockspeeds thing is just straight perplexing to me considering its an apples to oranges comparison outside of seeing the efficiency of the architecture. Understand to me, and seemingly many others, that you have to overclock the 7970 to reach the 680 speeds, so you are then comparing an overclocked card to a stock card. Thats not fair to me is all. I didnt find it fair when Nvidia had to clock up to reach AMD speeds either, so its a two way street.
I just want to clarify that I am not fired up about this subject either (nor does my post even seem like it). Again, I was just replying to what I thought you had a question about.
That said, I already conceded, in another thread you were discussing this same subject, that comparing top end overclocks is a valid comparison, but the matching clockspeeds thing is just straight perplexing to me considering its an apples to oranges comparison outside of seeing the efficiency of the architecture. Understand to me, and seemingly many others, that you have to overclock the 7970 to reach the 680 speeds, so you are then comparing an overclocked card to a stock card. Thats not fair to me is all. I didnt find it fair when Nvidia had to clock up to reach AMD speeds either, so its a two way street.
I just want to clarify that I am not fired up about this subject either (nor does my post even seem like it). Again, I was just replying to what I thought you had a question about.
If someone wants to test cards against eachother in what ever way the want,- they can, my opinion testing them against eachother clock for clock is valid, to be told is methods or my opinion is wrong by someone with a different opinion is just that, another opinion.....
He will have to live with my opinion because his opinion will not change mine no mater how hard he pushes it. The 1100T stock is 3.3Ghz the 1090T is 3.2Ghz, it means nothing, they are identical, they simply decided one is this clock and the other is that clock.. i decide its another clock!
Perhaps he should just relax and accept i have my own opinion.
@ ED, stock is just what they decided stock is, its not a measure of what the card is capable of, that's why overclocking matters to me.
Last edited:
- Joined
- Jun 29, 2005
- Location
- Over Here!
I think what everyone is saying is thank you for posting your benchmarks, I personally, understand the time you put into it, and appreciate the scope of what you did. However, I also see some suggestions on how your benchmarks could be better, that's all.
No one in here is posting/trolling, we're all after the same information; what they're saying is the manner in which the benchmarks were ran could of been "optimized" a little more... ie. put both cards at their respective factory/reference speeds and run some tests, then bump them up 10% and run some more etc...
I look forward to your future benchmarks, especially seeing how the 7970 scales!
No one in here is posting/trolling, we're all after the same information; what they're saying is the manner in which the benchmarks were ran could of been "optimized" a little more... ie. put both cards at their respective factory/reference speeds and run some tests, then bump them up 10% and run some more etc...
I look forward to your future benchmarks, especially seeing how the 7970 scales!
I do sincerely appreciate the effort and time he put in to this testing, and it is relevant as a comparison of two completely different architectures at the same clockspeed, no doubt.
Again man, just answering a question I thought you had, not trying to steer your opinion in any one way...
Now really, stop telling me to relax.
@ TnB - Can you run the same benchmarks at the same % overclock you clocked the 7970?
- Joined
- Jun 9, 2010
OP
- Joined
- Jan 19, 2006
- Location
- Florida
- Thread Starter
- #29
nice benches and comparison... but totally flawed methodology. I am not hating on you, just trying to help you do better next time.
You cannot overclock a 7970 and compare it to a stock 680. That is no way a fair and equal representation of the 680. clock for clock comparison cannot be made across architectures and companies (intel vs AMD)
As has been said by others, good comparisons are "stock vs stock" "stock+10% vs stock+10%" and "average OC vs average OC" (this would require a large sample size to get an accurate average since it is hit and miss how well cards OC)
"clock for clock comparison cannot be made across architectures"
This was what the benching was all about the architectures and to get a good feel of what both cards would do if they were at the same clocks! I know its somewhat flawed but is the best we could come up with
First I want to make it clear we are not here to BASH either AMD or NIVIDA this was just a simple test we decited to do for US and post our results.
Thanks bmwbaxter and well taken BUT all we were trying to see here is the difference in the way the architecture's run FPS at the same clock speeds nothing more we are going to OC them higher as the 680 should from what I understand do better when OCed.
I do like all the different opions here as it does give alot of insite from different prespectives which makes for some good info.
Thanks for the imput and lets see what higher OC brings and we may be doing 2X cards for scaling as well.
On a slight side note it was mentioned that the 680 will OC = BOOST we did do a mini load on both cards to make sure they were going to run at the same clocks as seen in the begining of the benching!
T&B
Last edited:
Hello, I think you got some interesting results.
I don't know if you do usually overclock, I would be curious to see a comparison with a maximum overclock, keeping in mind to have good temps and not too much noise.
I mean the level overclock one would usually do to not burn the cards within a few mounth, and to not have too much noise.
And I think the difference stays about the same. With a good custom cooler I think 7970 runs a bit faster.
I suppose the image given by the 2 cards is slightly different. I can distinguisch a printscreen of crysis 2 made with amd, and one made with nvidia cards.
With nvidia I see we can distinguisch a bit easyer the figure of another player and a different tesselation, while I like more the detail and colors that amd cards have. I think amd cards have a sharper image.
Would you like to post some printscreen pictures of the same or a similar image, one on top of the other, so that we can see how do these look?
About architecture, I've read, that nvidia begun to focus more on architecture and efficiency because the 28nm transistors are too expensive. So the effort of nvidia was in this direction.
Last question, what cpu do you have?
Tanks for the tests.
Generally speaking of tests, I wrote a tomshardware writer, I really would like to compare side by side two video of for example a game benchmark, or a similar action to tell differencies.
And not for last, with my 2x6870 in hd I get oft 80-100 fps on the second higher detail in crysis 2. Online I realized that in fast action I was loosing fps, for very short times, less than a second, but sufficient to get me killed. So an average of 80 fps wasn't enough!!
This reminds me to who said with the 500 series of nvidia, you cud get a 45-50 stable, while with similar priced amd cards you cud get 60-70 fps, but with the lovest points down to 20-30 fps.
That is absolutely not enough to play fast games.
And it's a result you can't always tell from numbers but when you see the video or better when you play you realize there it's something that makes the difference.
I don't know if you do usually overclock, I would be curious to see a comparison with a maximum overclock, keeping in mind to have good temps and not too much noise.
I mean the level overclock one would usually do to not burn the cards within a few mounth, and to not have too much noise.
And I think the difference stays about the same. With a good custom cooler I think 7970 runs a bit faster.
I suppose the image given by the 2 cards is slightly different. I can distinguisch a printscreen of crysis 2 made with amd, and one made with nvidia cards.
With nvidia I see we can distinguisch a bit easyer the figure of another player and a different tesselation, while I like more the detail and colors that amd cards have. I think amd cards have a sharper image.
Would you like to post some printscreen pictures of the same or a similar image, one on top of the other, so that we can see how do these look?
About architecture, I've read, that nvidia begun to focus more on architecture and efficiency because the 28nm transistors are too expensive. So the effort of nvidia was in this direction.
Last question, what cpu do you have?
Tanks for the tests.
Generally speaking of tests, I wrote a tomshardware writer, I really would like to compare side by side two video of for example a game benchmark, or a similar action to tell differencies.
And not for last, with my 2x6870 in hd I get oft 80-100 fps on the second higher detail in crysis 2. Online I realized that in fast action I was loosing fps, for very short times, less than a second, but sufficient to get me killed. So an average of 80 fps wasn't enough!!
This reminds me to who said with the 500 series of nvidia, you cud get a 45-50 stable, while with similar priced amd cards you cud get 60-70 fps, but with the lovest points down to 20-30 fps.
That is absolutely not enough to play fast games.
And it's a result you can't always tell from numbers but when you see the video or better when you play you realize there it's something that makes the difference.
OP
- Joined
- Jan 19, 2006
- Location
- Florida
- Thread Starter
- #31
Hello, I think you got some interesting results.
I don't know if you do usually overclock, I would be curious to see a comparison with a maximum overclock, keeping in mind to have good temps and not too much noise.
I mean the level overclock one would usually do to not burn the cards within a few mounth, and to not have too much noise.
And I think the difference stays about the same. With a good custom cooler I think 7970 runs a bit faster.
I suppose the image given by the 2 cards is slightly different. I can distinguisch a printscreen of crysis 2 made with amd, and one made with nvidia cards.
With nvidia I see we can distinguisch a bit easyer the figure of another player and a different tesselation, while I like more the detail and colors that amd cards have. I think amd cards have a sharper image.
Would you like to post some printscreen pictures of the same or a similar image, one on top of the other, so that we can see how do these look?
About architecture, I've read, that nvidia begun to focus more on architecture and efficiency because the 28nm transistors are too expensive. So the effort of nvidia was in this direction.
Last question, what cpu do you have?
Tanks for the tests.
Generally speaking of tests, I wrote a tomshardware writer, I really would like to compare side by side two video of for example a game benchmark, or a similar action to tell differencies.
And not for last, with my 2x6870 in hd I get oft 80-100 fps on the second higher detail in crysis 2. Online I realized that in fast action I was loosing fps, for very short times, less than a second, but sufficient to get me killed. So an average of 80 fps wasn't enough!!
This reminds me to who said with the 500 series of nvidia, you cud get a 45-50 stable, while with similar priced amd cards you cud get 60-70 fps, but with the lovest points down to 20-30 fps.
That is absolutely not enough to play fast games.
And it's a result you can't always tell from numbers but when you see the video or better when you play you realize there it's something that makes the difference.
The specs for the test are at the top of the post both systems were on a 3820.
I'm not a gamer but a bencher so I will have my friend answer the gaming aspects for you (abeeftec) but I can tell you that we did do OC runs and I havent had time to post them. As far as my friend goes he has told me he likes the ATI/AMD crads for gaming I beleive so lets have him answer. Will ost OC runs asap bit I can say there was no difference in the results from stock same clock runs.
T&B
Similar threads
