• 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) ***

Martini1991 said:
No benchmark shows a 4770K doubling an FX83's performance except AMD's blender example.
With my absolutely basic 4.4GHZ clock and absolute bog standard 1600MHZ CL9 RAM, using AMD's file, and the blender they link to, I'm doubling an FX83's performance at similar clocks.
Zen was quadrupling performance.
Click to expand...

No benchmark shows a 4770K doubling an FX83's performance except AMD's blender example.
Click to expand...

If thats the case then surely Blender is a far worse case for AMD than CineBench, be that as it may Zen's performance improvement is far greater than one would assume with CineBench doing the same thing as Blender.
 
humbug said:
If thats the case then surely Blender is a far worse case for AMD than CineBench, be that as it may Zen's performance improvement is far greater than one would assume with CineBench doing the same thing as Blender.
Click to expand...

That's a slant you're putting on it.
On Tom's Blender benchmark (Since someone's linked it) it's not close to doubling performance (4770K versus FX83)

If something's against the norm, the response should be to question, not to blindly follow it.
 
Martini1991 said:
That's a slant you're putting on it.
On Tom's Blender benchmark (Since someone's linked it) it's not close to doubling performance (4770K versus FX83)

If something's against the norm, the response should be to question, not to blindly follow it.
Click to expand...

One thing you need to know / remember, you can only get an accurate comparison between hardware if you render out the same image, Did Toms use AMD's Zen image?
A different image takes a different amount of time to render out, as its different.
It can't be used as a comparison :)
 
humbug said:
One thing you need to know / remember, you can only get an accurate comparison between hardware if you render out the same image, Did Toms use AMD Zen image?

It can't be used as a comparison :)
Click to expand...

It doesn't matter what image Tom's used as long as it was consistent between their hardware runs, in the benchmarks there's not close to the disparity between the FX83 and the i7 4770K as there is in AMD's benchmark. In Tom's benchmark (Which took ~4 minutes for the FX83, and just over 3 for the 4770K) there's ~25% between a 4770K and FX83, yet AMD's test shows me doubling an FX83's performance.

AMD have shown off a test in which their new CPU is 4 times faster at 3.4GHZ than 4.5GHZ FX83's. How can you NOT question that?
 
Last edited:
Martini1991 said:
It doesn't matter what image Tom's used as long as it was consistent between their hardware runs.

AMD have shown off a test in which their new CPU is 4 times faster at 3.4GHZ than 4.5GHZ FX83's. How can you NOT question that?
Click to expand...


Where did AMD test a 4.5Ghz FX83## in this event?

I'm going to guess that they didn't, that Toms used one in their test completely independently from this event using a generic image, IE not the same image AMD used in this event, different images rendered out on the same system running all the same hardware take different times as the image is different.
 
humbug said:
Where did AMD test a 4.5Ghz FX83## in this event?

I'm going to guess that they didn't, that Toms used one in their test completely independently for this event using a generic image, IE not the same image AMD used in this event, different images rendered out on the same system running all the same hardware take different times as the image is different.
Click to expand...

AMD didn't.
But they've linked their image to use in Blender to test (And a link to blender), which people have done.

I need to go to sleep, the lack of logic is killing me.
 
I just ran the same benchmark in blender 2.71 and got 1 minute 58.

Clearly there are some performance regressions with the newer version of Blender on AMD FX processors.
 
Martini1991 said:
AMD didn't.
But they've linked their image to use in Blender to test, which people have done.

I need to go to sleep, the lack of logic is killing me.
Click to expand...

Where is this link you speak of? a simple question would be did Toms use AMD's Image from this event? if they didn't the comparison is flawed.
 
The Ryzen name will take some getting used to it sounds like some sort of kitchen cleaning product.

I'm still suspicious that AMD are using a custom made demo/Blender for benchmarking, it's bound to be optimised to portray the new architecture in the best light possible and if that is a 3.4ghz+ Zen matching a 3.2ghz 6900K in a handpicked test then what happened to the faster IPC?

I have a feeling AMD are setting themselves up for a fall here because it's real world applications and benchmarks that really matter, like others have said they should have just run Cinebench or something and we'd have had a much clearer idea of real world performance, it seems like they have something to hide.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom