I would just like to know where you base that conclusion to ?? 980 was released when Sandybridge was on the market.
Still was a better product at its time. AMD only got worse year after year.
I would just like to know where you base that conclusion to ?? 980 was released when Sandybridge was on the market.
And you think that the FX8350 closed the gap instead of it being expanded?
2500K was around 36½% faster in ST than 980BE using Cinebench.
2500K was around 43% faster in ST than 1100T using Cinebench.
4690K is around 60½% faster in ST than 8350 using Cinebench.
And you think that the FX8350 closed the gap instead of it being expanded?
2500K was around 36½% faster in ST than 980BE using Cinebench.
2500K was around 43% faster in ST than 1100T using Cinebench.
4690K is around 60½% faster in ST than 8350 using Cinebench.
Not sure if serious, even if you take Cinebench as a metric you clearly dont use the Single Thread performance of a MultiThreaded/Multi processor rendering Application
And something else,
Core i5 2500K was released on January 2011, 980BE was released on May 2011, that is after the Core i5 2500K
FX8350 was released in October 2012 when IvyBridge was in the market. Core i5 4690K was released on July 2014.
Get your facts strait.
And if you didnt get the memo yet. "Moar cores" doesnt sell.
And yet you are using a MutliThread/Multi CPU Rendering Application.
Whatever floats your boat.
No one benchmark is perfect, it always helps to have as many as one can, but if absolutely forced to choose just one benchmark, I struggle to think what is a better representative of both Single Thread and Multi Thread performance, than Cinebench.
If you have some better alternatives, I would be happy to hear about them.
No one benchmark is perfect, it always helps to have as many as one can, but if absolutely forced to choose just one benchmark, I struggle to think what is a better representative of both Single Thread and Multi Thread performance, than Cinebench.
If you have some better alternatives, I would be happy to hear about them.
Ah, the old "Construction cores are inferior to K10.5" trope. Tell you what.
The 980 was a Black Edition chip, meant for overclocking, though in all fairness, it was close to its realistic clockspeed wall @ stock anyway. Very little headroom except maybe if you put it on the right AM3+ board and dump volts on it, and even still, that trick worked better on Zosma/Thuban than Deneb. Even still . . . it's a BE.
In the interest of friendly discourse, let's pit my 4.7 ghz A10-7700k against pretty much any%2
It is not about the benchmark, Cinebench is fine for Single and Multi. It is how he manipulated the benchmark, he only used a single thread performance on a highly Intel optimized benchmark to come to a conclusion that suited his agenda of making the AMD products look worse in a single non real world use scenario. That is, nobody will ever use Cinema 4D to render with a single core.
He knew that if he would post the Cinebench Multithread performance or any other application, his claim would be dismissed instantly.
As for other benchmarks, the first to come in mind is Geekbench that has both single and Multi.
It was the top x4 chip offered. Just like the a10-7850k is the top chip offered. (You could use the FX-4350 and get similar results)
CB is being used as a benchmark of ST peformance. The fact that it is rendering is irrelevant; its the performance of a single core.
Likewise GB is useless as EVERY SINGLE TEST is pretty much perfectly parallelizable.
We can use multicore maya rendering.
2500k vs. 980 = 11% faster
3570k vs. a10-5800k = 56% faster
4670k vs. a10-7850k = 75% faster
Its gotten so bad that AMD's quads are now compared to intel's dual cores.
This illustrates it better.
Performance for quad core is stagnant since PII 980.
2500k -> 4670k 21% gain
980 -> a10-7850k 8% loss
AMD has launched 8 core chips though. Prices have changed. Nonetheless performance per core has not increased.
AMD sells them as quadcores. Are you saying AMD is being...untrue?
Quite a sudden change for you. Or is it just a new goalpost shift in an attempt to avoid having to accept the realities?
AMD sells 2M4T chips as quadcores.
Obviously, those are SMT Cores.Intel sells 2C/4T chips as dualcores.
I assume we should only compare FX8xxx to i7s now?
It also sells 4M8T as Okta-Cores. so ?? I believe you know very well what CMT is. So dont try to be smart here with me.
Obviously, those are SMT Cores.
If you want to compare them against Core i3 be my quest
Why should we compare them to i3? Not that the single threaded part will be any different. AMD have defined them as quadcores. So the compare must be to previous AMD quadcores and to Intel quadcores.
Nomatter how you try and twist, complain and shift goalposts. AMD had better products with the 980BE, 1100T etc than they do now.
For those that want to be correct, one module should be compared against one SMT Core.
So you are saying AMD is wrong, just so you can define whatever testcase you wish to avoid having to face reality. You could just have said that to begin with.
It was the top x4 chip offered. Just like the a10-7850k is the top chip offered. (You could use the FX-4350 and get similar results)
CB is being used as a benchmark of ST peformance. The fact that it is rendering is irrelevant; its the performance of a single core.
Likewise GB is useless as EVERY SINGLE TEST is pretty much perfectly parallelizable.
We can use multicore maya rendering.
2500k vs. 980 = 11% faster
3570k vs. a10-5800k = 56% faster
4670k vs. a10-7850k = 75% faster
Its gotten so bad that AMD's quads are now compared to intel's dual cores.
This illustrates it better.
Performance for quad core is stagnant since PII 980.
2500k -> 4670k 21% gain
980 -> a10-7850k 8% loss
AMD has launched 8 core chips though. Prices have changed. Nonetheless performance per core has not increased.