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

~>Dg<~ · 14 Dec 2016 at 22:33

until we see proper benchmarks of popular games and programs the wild speculation is almost hysterical.

Martini1991 · 14 Dec 2016 at 22:33

jagga1968 said:
dont forget how you were trying to pour water on the yesterdays benchmarks.

now the tune is changing.

I'm still as sceptical as ever about Blender and the validity of those results.
No tune's changing. It's called being objective.

To me you're just some AMD fanatic who's wandered in to troll, and that's how I'll see you.

Harlequin · 14 Dec 2016 at 22:34

~>Dg<~ said:
until we see proper benchmarks of popular games and programs the wild speculation is almost hysterical.

explain how a live screens of blender running is not a proper benchmark?

or is your x99 too much salt

edit:

martini - already showed you tomshardware showing the disparity from 2015.

Martini1991 · 14 Dec 2016 at 22:34

bru said:
Also if this leaked pricing is correct, $500-$150 for Ryzen where on earth do all the Bristol ridge APU's fit in. Obviously $150 downwards.

Actually I've just looked at the currant APU line up, from £129 downwards, so maybe it might work.

We can't really do any comparisons with sterling.
Thanks Brexit.

jagga1968 · 14 Dec 2016 at 22:35

if amd have learned any lessons from fury x launch they wont make the same mistake.

jagga1968 · 14 Dec 2016 at 22:36

Martini1991 said:
I'm still as sceptical as ever about Blender and the validity of those results.
No tune's changing. It's called being objective.

To me you're just some AMD fanatic who's wandered in to troll, and that's how I'll see you.

amd fanatics wouldnt buy sandybridge and a 6700k.

pot calling the kettle black again.

Martini1991 · 14 Dec 2016 at 22:37

Harlequin said:
explain how a live **** of blender running is not a proper benchmark?

or is your x99 too much salt

Because my 4770K is twice as fast as an FX8300 if we use Blender to compare them using the same version and same image.
An i7 4700K isn't twice as fast as an FX8300 in stuff like Cinebench (Which is apparently just the same......) So it should be questioned.

When people start throwing in "salty" and the like, I don't think rational discussion works.

Harlequin · 14 Dec 2016 at 22:39

and again , I have already shown that in 2015 , using a previous version of blender - the 4770k is twice as fast as the 8320

Martini1991 · 14 Dec 2016 at 22:39

Blender's obviously a crap benchmark if it shows a ridiculously inflated performance difference between two CPU's that's not anywhere else.

Harlequin said:
and again , I have already shown that in 2015 , using a previous version of blender - the 4770k is twice as fast as the 8320

That doesn't matter at all.
It just means blender shouldn't be used at all as flipping benchmark.

jagga1968 said:
amd fanatics wouldnt buy sandybridge and a 6700k.

pot calling the kettle black again.

Even AMD fanatics don't want to buy second best.

jagga1968 · 14 Dec 2016 at 22:40

Martini1991 said:
Because my 4770K is twice as fast as an FX8300 if we use Blender to compare them using the same version and same image.
An i7 4700K isn't twice as fast as an FX8300 in stuff like Cinebench (Which is apparently just the same......) So it should be questioned.

When people start throwing in "salty" and the like, I don't think rational discussion works.

unless you can say for certain that the same settings were used by you and amd benchmark done yesterday your theory is meaningless.

SiDeards73 · 14 Dec 2016 at 22:40

This is a decent thread from AMD Reddit where one guy got people to benchmark blender with that RyZen benchmark

https://www.reddit.com/r/Amd/comments/5i6w0w/submit_your_blender_test_for_comparison/

Makes for interesting reading

Martini1991 · 14 Dec 2016 at 22:42

jagga1968 said:
unless you can say for certain that the same settings were used by you and amd benchmark done yesterday your theory is meaningless.

It was people here.
AMD linked to the benchmark and cinebench version.

If I started calling my 4700K twice as fast as an FX83 2 weeks ago I'd be attacked.

Now suddenly I'm attacked for saying an FX8300 isn't half the speed of a 4700k.

Cool.

Harlequin · 14 Dec 2016 at 22:42

Martini1991 said:
That doesn't matter at all.
It just means blender shouldn't be used at all as flipping benchmark.

9 pages of you harping on about blender ; the software is used by professionals and has the ability to compare cpu`s using the same image.

by definition , that's a benchmark. you may or may not like it but that does not matter at all. The point is , the software can be used to level the playing field.

edit: you have provided no proof at all from your claim. Supply proof. as I have.

SiDeards73 · 14 Dec 2016 at 22:42

Martini1991 said:
Even AMD fanatics don't want to buy second best.

I admit I looked at the 8350 when I swapped from my 2500k to 4770k, but the performance was just not there.

I firmly believe though that RyZen will match my 4770k at a minimum while offering more cores, so it's a winner for me.personally. who doesn't like knew toys?

Martini1991 · 14 Dec 2016 at 22:43

Harlequin said:
9 pages of you harping on about blender ; the software is used by professionals and has the ability to compare cpu`s using the same image.

by definition , that's a benchmark. you may or may not like it but that does not matter at all. The point is , the software can be used to level the playing field.

If these results were used 2 weeks ago, one would be called out for cherry picking to make an FX8300 look bad.

Amazing.

Harlequin · 14 Dec 2016 at 22:44

Martini1991 said:
If these results were used 2 weeks ago, one would be called out for cherry picking to make an FX8300 look bad.

Amazing.

Again you supply no proof. Just an opinion.

Martini1991 · 14 Dec 2016 at 22:45

Okay.
4700K is twice as fast as an FX8300.

Awesome.

I honestly can't believe what I'm reading.

People try and argue so much how the FX83's put up a decent fight in heavy threaded situations, but here we go, trying to now argue the opposite and how the i7 4700k is twice as fast.

Martini1991 · 14 Dec 2016 at 22:46

SiDeards73 said:
I admit I looked at the 8350 when I swapped from my 2500k to 4770k, but the performance was just not there.

I firmly believe though that RyZen will match my 4770k at a minimum while offering more cores, so it's a winner for me.personally. who doesn't like knew toys? ;)

There's no way that Ryzen flagship isn't in a league of its own compared to the 4770K

~>Dg<~ · 14 Dec 2016 at 22:46

Harlequin said:
explain how a live screens of blender running is not a proper benchmark?

or is your x99 too much salt

edit:

martini - already showed you tomshardware showing the disparity from 2015.

blender the prog that means all. :rolleyes:

seriously theey choose benchmarks that favour the cpus outcome. why did they not show game benchmarks ?

if the product was that strong they would.all they could muster was a blender benchmark.if you cant see that as iffy then i dont know what to say.

some in here are expecting a chip for 300 quid to beat a 1000 quid chip.no.maybe in a couple of progs no one uses.thats why they chose them.is that not obvious.

if for eg its 40 percent better than older 8 core.you know where that puts it ?

slower than a 6700k slower than 5820k by quite some bit.so no im not worried or bothered.

im just slightly bemused at why some think amd are going to pull out a 1000 quid cpu killer for peanuts.it aint happening.

jagga1968 · 14 Dec 2016 at 22:47

Martini1991 said:
Okay.
4700K is twice as fast as an FX8300.

Awesome.

ffs give it a rest

Competitor rules

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

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