• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*** AMD "Zen" thread (inc AM4/APU discussion) ***

Blender benchmark.
I've seen blender used in education :p.

But it doesn't change anything, people wouldn't have put stock in it before because of the ludicrous performance disparity between Intel and AMD. A disparity that isnt anywhere else.

I'm not sure how it can be portrayed as any form of fanboyism as I'm arguing that an FX83 is faster in comparison to an i7 than blender would have you believe.
 
Last edited:
~>Dg<~ said:
i think we mean in general.dont we.

not many as a perecentage will use blender.

what most who are probably ready for upgrading want to see is popular new game benchmarks.editing programs people use a lot.encoding progs people use a lot.

not random things that the majority of us wont use.

when pretty obscure benchmarks are shown it often tells you why in itself.

if it was as fast as the 6900 in games for eg why would you not show this ? that would be massive hype and there would be queues forming to get them.but nope just blender.everyones fav prog :D
Click to expand...

Games are by and large not CPU limited, showing off Zen to the public using game benchmarks would have been utterly ridiculous. The games they did show which weren't talked about score wise were really just to show it running the latest game without problem, but benchmarking said game in relevance to the CPU itself is near worthless.

If they benchmarked in a CPU limited fashion which would mean say 720p and low res, you'd all ask why they ran it so low res and point out it's not relevant to real usage.

Blender isn't obscure, it's a real program that is tough on cpu performance and used by a lot of people. The $500 8 core 16 thread chip isn't going to immediately appeal to the mass market of general gamers anyway. The massive majority of gamers do NOT buy Intel's 6 to 10 core CPU only chips, the massive majority of gamers buy something between £50-250. People who buy more power mostly buy it because they are A, an enthusiast who just wants something faster even if they won't necessarily use that speed at all or B, someone who actually runs heavily cpu limited programs... like Blender.

It's also an early look, it wasn't a launch, there weren't 48 things benchmarked, and shock horror, they picked something that showed it in a good light as well as being appropriate for the main people who would buy such a chip.

When it comes to a launch you'll get more benchmarks, both from AMD and from review sites. If that was a launch event the constant whining about the choice of benchmark would make sense, but it's an early look, it isn't intended to be in depth but people are complaining because it wasn't in depth. Intel would never release a full suite of benchmark results before launch, neither would Nvidia for anything they sell, but AMD apparently have to?
 
I don't understand why this argument that because the FX series CPU's do better against Intel in CineBench than in Blender it must mean Zen matching Intel in Blender is fake. For one the FX CPU's and Zen are very very very different CPU's. they are not comparable, Zen is actually far more comparable to Intel's CPU's, that right there is probably the answer to that question.

It's a very bad idea for AMD to fake this, if it turns out that infact Zen is much slower than Intel in Blender AMD would be in for a lawsuit and what little trust remains will be gone, it would spell the end of AMD.

AMD have done some daft things in their time, but to fake something, like this, on this scale? that's a conspiracy much too far.
 
Last edited:
I'm not saying they've faked the benchmark.
I just have no stock in the results of a benchmark which can produce such disparity that exists no where else.

Seemingly the only reason people are even paying attention this benchmark all of a sudden is because it's made AMD Zen look good.

I don't doubt Zen will be good, the handbrake did more than enough to convince me of that. But my problem is how blender would have been ridiculed had people used it previously as a benchmark, because of how poor its made the Current AMD stuff look. The reason (rightfully so) is that how can people trust a benchmark which displays disparity that doesn't exist anywhere else.
 
Last edited:
I'm not sure how the Blender application operates but my concern would be that the demo file provided by AMD favours the Zen architecture at the expense of others, that's why I'd have liked to see a stock benchmark that is a known quantity used rather than something AMD themselves have cooked up.

Showcasing Zen in a best case scenario under artificially chosen conditions is not technically cheating, even if it's not reflective of normal everyday conditions.
 
Last edited:
To put it another way.
If the i7 can double an FX83 in blender, then surely that means the i5 4670k is faster in blender than an FX83

You've got a hard time convincing people that an i5 is faster than an FX83 in games. Let alone going 100% it can best an FX83 in a heavily threaded benchmark.

You'd put no stock in an result which shows an FX83 getting outmuscled by a decent margin in such a circumstance.
 
I have never compared my FX-8 CPU's to the 4690K but in terms of performance i didn't notice any diffrence...

Anyway.... my 4690K @ 4.5Ghz / 4.2Ghz UnCore / 2400Mhz 16GB CL11

Edit 150 samples, thats what AMD ran.

1:26.23

3.4Ghz 8 Core Zen is nearly 3x faster, impressive.

ryzen.png
 
Last edited:
humbug said:
Anyway.... my 4690K @ 4.5Ghz / 4.2Ghz UnCore / 2400Mhz 16GB CL11

Edit 150 samples, thats what AMD ran.

1:26.23

3.4Ghz 8 Core Zen is nearly 3x faster, impressive.
Click to expand...

That's running a Blender demo file provided by AMD specifically to showcase Zen though? in Cinebench and other generic benchmarks it may only be 1.5x faster and that's my point. For me the benchmark AMD showed leaves more questions than it answers.
 
mmj_uk said:
That's running a Blender demo file provided by AMD specifically to showcase Zen though? in Cinebench and other generic benchmarks it may only be 1.5x faster and that's my point. For me the benchmark AMD showed leaves more questions than it answers.
Click to expand...

More conspiracy....

Its rendering polly's and tessellation triangles, there isn't anyway you can cheat that, the hardware just does what it does, it does not know what logo is on the file. No one who knows this stuff is asking these questions, and this is all over the 2D / 3D community forums, including Blenders own, its not a viable theory.

Again this is the sort of conspiracy theory that would be discovered the instance the chip is in anyone's hands other than AMD.

Tinfoil hating too far. :)
 
Last edited:
some of you guys need to settle down regarding cinebench, it can't be trusted when running on non intel cpus and there should not even be a debate about this.

cinebench is fine for intel to intel based comparison only. it will be fun to run cinebench on ryzen but it can't be used to make a purchasing decision on a non intel based cpu.

posted by someone whom has only ever owned one AMD CPU in past 40 years.
 
Looks like Zen is shaping up nicely. I'm looking forward to release / proper reviews

However, I'm firmly keeping my feet on the ground. I'm hoping it will be a touch worse than intel's offerings, and be a touch cheaper. If it works out any better than that, then I'll be pleased to be pleasantly surprised, and worse, then good job on the troll AMD, good job.

Fingers crossed things work out. It will be strange having a viable option to suggest again for gaming machines instead of just i5s / i7s
 
Ok, I know this arguing can be fun and all but could we all just take a deep breath and see if there is anything else we can actually work out that is meaningful or when the next news that will tell us something is going to drop?

We saw AMD launch an AMD product. It is safe to assume that they will have done that in a way that benefits them. They probably didn't outright lie about it and what they showed makes Ryzen look like a really compelling competitor to Intel at the enthusiast/prosumer level for the first time in quite a while.

The rendering benchmarks imply that in certain scenarios AMD have managed to get their IPC to very similar performance to Skylake. For them to be at that level in any main stream application is a massive achievement that I don't think anyone seriously expected.

It doesn't matter which benchmarks they used - we could always pick holes in any two or three as in this kind of show they were never going to show a complete test. Even if they did it would have been AMD doing it, so we would always doubt it was real.

Do we know when the next drip of information is going to drop? The train has definitely left the station, I am even ready to admit I am a passenger.
 
Beren said:
Ok, I know this arguing can be fun and all but could we all just take a deep breath and see if there is anything else we can actually work out that is meaningful or when the next news that will tell us something is going to drop?

We saw AMD launch an AMD product. It is safe to assume that they will have done that in a way that benefits them. They probably didn't outright lie about it and what they showed makes Ryzen look like a really compelling competitor to Intel at the enthusiast/prosumer level for the first time in quite a while.

The rendering benchmarks imply that in certain scenarios AMD have managed to get their IPC to very similar performance to Skylake. For them to be at that level in any main stream application is a massive achievement that I don't think anyone seriously expected.

It doesn't matter which benchmarks they used - we could always pick holes in any two or three as in this kind of show they were never going to show a complete test. Even if they did it would have been AMD doing it, so we would always doubt it was real.

Do we know when the next drip of information is going to drop? The train has definitely left the station, I am even ready to admit I am a passenger.
Click to expand...

I wouldn't expect any other real information until NDA drops on release reviews TBH

You will get your usual wccftech type "news" of course, if that's your bag.
 
new boy said:
I wouldn't expect any other real information until NDA drops on release reviews TBH

You will get your usual wccftech type "news" of course, if that's your bag.
Click to expand...
Meh, no - but I guess it will have to do. I don't know why I am so excited - I was genuinely excited about the 480 launch too.

I know it is Q1 - do we know when we will have a definite launch date even?
 
Beren said:
Meh, no - but I guess it will have to do. I don't know why I am so excited - I was genuinely excited about the 480 launch too.

I know it is Q1 - do we know when we will have a definite launch date even?
Click to expand...

I was fully emotionally invested in the 480 launch. Sold my 290x in anticipation and everything lol. I went a bit too crazy and go caught in the hype. Not this time! Feet. firmly. on. the. ground.
 
You must log in or register to reply here.
Back
Top Bottom