Official AMD Ryzen Benchmarks, Reviews, Prices, and Discussion

[Glo.]

I fear that may get a bit annoying - AMD clearly now have the technology to produce a really good single chip gaming solution, whether they've got the spare R&D for a niche thing like that is another matter.

The APU in question is called Horned Owl. It is not meant to be for gaming market, first, despite its possibilities being used there.

There is a little confusion. AMD will offer Ryzen 3 CPUs in 2h of 2017. But the Raven Ridge APUs are not going to be available for mainstream desktop market till 2018. On the other hand, the Horned Owl chip is going to be available 2h 2017.

Horned Owl was designed first in mind of server, machine learning and Embedded markets. I thought that it will be called Raven Ridge.

 

[lolfail9001]

Horned Owl was designed first in mind of server, machine learning and Embedded markets. I thought that it will be called Raven Ridge.

It is Raven Ridge, same way Zeppelin IS Summit Ridge.

On the other hand, the Horned Owl chip is going to be available 2h 2017.

Laptop Raven Ridge, any issues? From what little i see on both, Horned Owl is basically a big laptop/embedded APU, while Banded Kestrel is an attempt at 15W ultrabooks.

 

[Glo.]

It is Raven Ridge, same way Zeppelin IS Summit Ridge.

Laptop Raven Ridge, any issues? From what little i see on both, Horned Owl is basically a big laptop/embedded APU, while Banded Kestrel is an attempt at 15W ultrabooks.

Yep, and Banded Kestrel has 5.5 CU's.

Great job.

 

[CatMerc]

The 1400 and 1500X are too closely priced for what they offer - AMD should have priced the 1400 at ~130 and it would have killed off the i3s once and for all.

They are probably going to do that with Raven Ridge I supppose.

They probably want to leave room for R3's. Remember they still have Bristol Ridge APU's to sell, so they might be uncomfortable to sell R3's below 100$.

 

[cytg111]

they are all unlocked so I say it's perfectly OK as it is. OCers will choose the cheaper one everybody else the 1600x. No need for a 3rd sku.

The quads are probably the worst binning chips that can't reach higher clocks (and not necessarily dies with broken cores).

And we should all pause and appreciate the fact that these chips is churning out of glofos ovens... to my uneducated guesstimation that was by far the biggest hurdle. Its almost a miracle.

 

[Glo.]

They probably want to leave room for R3's. Remember they still have Bristol Ridge APU's to sell, so they might be uncomfortable to sell R3's below 100$.

Yep, and next year the Raven Ridge APUs are supposed to replace Bristol Ridge in this market.

 

[lolfail9001]

Yep, and Banded Kestrel has 5.5 CU's.

Why not 6, if we have established that Horned Owl is simply a 12 CU with 1 CU disabled (and it does not make sense otherwise, does it?).

 

[Glo.]

Why not 6, if we have established that Horned Owl is simply a 12 CU with 1 CU disabled (and it does not make sense otherwise, does it?).

Let me put this this way. NOTHING currently about the APUs makes sense.

 

[lolfail9001]

Let me put this this way. NOTHING currently about the APUs makes sense.

Strictly speaking we do not exactly know anything about them either, except their code names and Kestrel being the smaller one.

 

[Puffnstuff]

I think we'd better wait for them to be released and tested by reputable sources before we pass judgement on any of the new parts.

 

[Whitestar127]

Hardware Unboxed just did some testing on the 2500K. The results go directly against what the Adored TV "Ryzen - the tech press loses the plot" video claimed.

Thought this was relevant, given the controversy sorrunding the reveiw process. Doesn't have a whole lot to do with Ryzen itself though.

 

[Despoiler]

The timing of the AMD Windows scheduler announcement is to be expected. MS is in bug fixing & polish mode for the Creators 2 update. Both the fast and slow ring have builds of it. They won't be adding anything new until it goes GA.

 

[CatMerc]

Hardware Unboxed just did some testing on the 2500K. The results go directly against what the Adored TV "Ryzen - the tech press loses the plot" video claimed.

Thought this was relevant, given the controversy sorrunding the reveiw process. Doesn't have a whole lot to do with Ryzen itself though.

Entirely depends on the set of games.

Computerbase's testing used a different set of games, so they got different overall results.

 

[unseenmorbidity]

Entirely depends on the set of games.

Computerbase's testing used a different set of games, so they got different overall results.

The point didn't depend on which was faster anyways. The point was that the old slow 8 thread CPU made up a lot of ground. Whether it arguably passed it or not is irrelevant.

These techtubers are having a pissing contest atm, which would be comical, if it wasn't so pathetic.

 

[Insomniator]

Have they even caught up though? Those results aren't even close, and that looked like a pretty good set of modern games and the battlefield results directly contradict each other. Someone is making **** up or there are other factors (like memory speed).

 

[USER8000]

Have they even caught up though? Those results aren't even close, and that looked like a pretty good set of modern games and the battlefield results directly contradict each other. Someone is making **** up or there are other factors (like memory speed).

They looked at the Battlefield 1 multiplayer and showed in areas with more people the Core i5 2500K was faster and in places with less people the FX8370 the opposite happened.

 

[tamz_msc]

One cannot take the HW Unboxed and Computerbase data and use them to refute one against the other. It will obviously depend on what games are tested, as well as the test methodology.

 

[Mockingbird]

Here's an interesting part:

"In official information for channel partners AMD revealed that Ryzen 5 are focused on 1080p gaming:

We’re thrilled to let you know we have an incredibly strong roadmap and our upcoming AMD Ryzen™ 5 processor will be focused on 1080p game play. The best is yet to come!"

So far Ryzen has not done as well in 1080p gaming.

 

[Whitestar127]

AdoredTV just put up their Ryzen review:

 

[dark zero]

Don't know if this deserves their own topic or not...

Seems that ZEN wil get a refresh on 2018....

but... Why Stoney Ridge is still alive at 2018????
@NostaSeronx is right?
And the refresh will be at 22nm?

 

[lolfail9001]

Here's an interesting part:

"In official information for channel partners AMD revealed that Ryzen 5 are focused on 1080p gaming:

We’re thrilled to let you know we have an incredibly strong roadmap and our upcoming AMD Ryzen™ 5 processor will be focused on 1080p game play. The best is yet to come!"

So far Ryzen has not done as well in 1080p gaming.

Plot twist: they mean GPU limited 1080p gaming and will use rx480 for official(tm) comparisons. In fact, i bet on the latter part.

 

[Malogeek]

How is an APU still an APU with no GPU cores? Isn't it just a CPU then?

edit: I think it was just to save room in the slide

 

[unseenmorbidity]

Plot twist: they mean GPU limited 1080p gaming and will use rx480 for official(tm) comparisons. In fact, i bet on the latter part.

That is probably accurate. They are likely calling it a 1080p gaming CPU, because it's cheap.

Which they are! $170 for an unlocked i7 is some good value! You can't even buy a used 4770k for that!

No CCX bugs either...

 

[.vodka]

I find it interesting that there's no mention of Zen 2 for Pinnacle Ridge. I mean, stuff like that should show up on the roadmap, right?

Hm, is this a refresh like Kaveri -> Godavari and Carrizo -> Bristol Ridge? Higher clock speeds with little tweaks here and there resulting in a polished product?

 

[beginner99]

The timing of the AMD Windows scheduler announcement is to be expected. MS is in bug fixing & polish mode for the Creators 2 update. Both the fast and slow ring have builds of it. They won't be adding anything new until it goes GA.

This is typical AMD and on some part Bulldozer all over again. I mean their design decision with the ccx was made years back (before win 10 release...) and ES version were at least around for a year. So there should have been more than enough time to fix it. Same as with bulldozer.

As the underdog you can't really on software getting adjusted to your uArch. Won't happen. And just denying the issue in PR statements doesn't help either. I really wonder why AMD never learns. Intel can just do nothing and watch AMD dig their own grave. I mean they now made an offical statement. Any backtracking would be admitting huge incompetence.