New Macbook Pro with vega graphics coming soon

[wiregen]

People keep wondering what the performance will be but there's already a "vega" 20 and 24 out on the market...

They're in the Intel NUC 8s...
Base model has a RX Vega M GL with a vega chip, 20 CUs and RX Vega M GH with 24 CUs

Both got a 4 GB single stack of HBM2 as well.

Only difference is that the boost clock speed is a bit lower at around 1200 mhz.

Since they perform at between a 1050 and a 1050 TI, I fully expect these Vega 20 chips to perform at about the same level as the 1050ti and Vega 16 to be at about a 1050.

What is the 560X comparible in the nvidia relm?

 

[Thysanoptera]

No there arent any...

Vega 56 is 500 mm2 die sized monster. Vega 20 is under 180 mm2. HOW THEY CAN BE THE SAME CHIP?!

It is completely different silicon. Completely different Circuit.

Easy, I just pointed out that what you included as differences consisted of cutting down on number of active components, which is what the silicon makers do when they have crappy yields - thus I said it is a definition of binning. Of course it is physically a different chip. I'm genuinely curious if there is anything more to it due to "mobile first philosophy".

 

[_Kiki_]

I don't really see how this is relevant at all given that the MBP is not going to use Kaby Lake G or Vega M...

think different... again....

 

[Thysanoptera]

Yes, there are. AMD claims that Vega Pro 16/20 comes with "optimised geometry engine" compared to the desktop Vega. The chip itself is the Vega Codename 12 that was leaked some time ago, while desktop 56/64 are Vega 10. All these numbers are rather confusing.

So I typed - vega "codename 12" - in google and got unexpected, although pretty results. I give up Let's wait a month and see how this 'optimized geometry engine' plays out in real life.

 

[ondert]

I just got excited thinking it might be the expected 7nm Vega, however it seems to be the opposite.

 

[koyoot]

Easy, I just pointed out that what you included as differences consisted of cutting down on number of active components, which is what the silicon makers do when they have crappy yields - thus I said it is a definition of binning. Of course it is physically a different chip. I'm genuinely curious if there is anything more to it due to "mobile first philosophy".

You clearly have no idea what binning is.

Binned chips are for example golden samples of i7-8700K which are clocking to 5.3 Ghz on 1.3v. Those are best binned chips.

What you have described is PRODUCT SEGMENTATION. It has nothing to do with this case, because Vega 12 and Vega 10 are two, completely different dies, different circuits, different designs.

Its like saying that Polaris 11 is cut down version of Polaris 10. Which is not true.

 

[leman]

People keep wondering what the performance will be but there's already a "vega" 20 and 24 out on the market...

Welcome to the very confusing world of GPU marketing.

The GPUs you refer to are based on the chip called "Vega M" which is a hybrid on basis of AMD Polaris with some components cut out from AMD Vega elements. This is not a Vega chip. They don't have much to do with the new Vega Mobile 16 and 20 since they use different core architecture.

think different... again....

This is not "riddle daily". I am not the only one that doesn't understand what point you are trying to make. We would appreciate if you explain it in more detail.

 

[Thysanoptera]

You clearly have no idea what binning is.

Binned chips are for example golden samples of i7-8700K which are clocking to 5.3 Ghz on 1.3v. Those are best binned chips.

What you have described is PRODUCT SEGMENTATION. It has nothing to do with this case, because Vega 12 and Vega 10 are two, completely different dies, different circuits, different designs.

Its like saying that Polaris 11 is cut down version of Polaris 10. Which is not true.

If you want to argue about semantics feel free to edit wikipedia on definition of product binning:

Defects in manufacturing are not always fatal, however; in many cases it is possible to salvage part of a failed batch of integrated circuits by modifying performance characteristics. For example, by reducing the clock frequency or disabling non-critical parts that are defective, the parts can be sold at a lower price, fulfilling the needs of lower-end market segments.

Welcome to the very confusing world of GPU marketing.

And Vega 20 was supposed to be the new 7nm chip.

 

[MBX]

Kinda lame of Apple to not have put the Vega in the MBP 2018 release 2-3 months ago. Annoying for those of us who just bought the 2018 MBP.

 

[koyoot]

If you want to argue about semantics feel free to edit wikipedia on definition of product binning:

Defects in manufacturing are not always fatal, however; in many cases it is possible to salvage part of a failed batch of integrated circuits by modifying performance characteristics. For example, by reducing the clock frequency or disabling non-critical parts that are defective, the parts can be sold at a lower price, fulfilling the needs of lower-end market segments.

And how does what you have described play with the idea that under 180 mm2 die is different bin of 510 mm2 die sized GPU?

It has nothing to do with it.

You can say that different bins of the same die are Vega 56 and Vega 64. Vega 20 is not different bin of that die, because it is different die, different design.

 

[leman]

And Vega 20 was supposed to be the new 7nm chip.

Vega mobile/pro 20 is not Vega 20. It’s Vega 12. As I said, it’s very confusing. One is name of the product, another is the Codename of the chip.

 

[koyoot]

Vega mobile/pro 20 is not Vega 20. It’s Vega 12. As I said, it’s very confusing.

Vega 10/12 is the design name. Vega 64/56 or Vega Pro 20 is the product name.

 

[leman]

Vega 10/12 is the design name. Vega 64/56 or Vega Pro 20 is the product name.

Yes, your terminology is better

 

[Thysanoptera]

And how does what you have described play with the idea that under 180 mm2 die is different bin of 510 mm2 die sized GPU?

And I will quote myself again, this time making the part that you missed more visible:

Easy, I just pointed out that what you included as differences consisted of cutting down on number of active components, which is what the silicon makers do when they have crappy yields - thus I said it is a definition of binning. Of course it is physically a different chip. I'm genuinely curious if there is anything more to it due to "mobile first philosophy".

Do you want to make another round?

 

[koyoot]

Yes, your terminology is better

Its not mine, but AMD's

To be precise: Vega/Polaris - codename of the architecture. Number after that indicates which GPU was first designed. 10 is first, 11 after it, 12 after 11. Thats all there is.

Navi 10 for example will be either Radeon RX 660 or 680 depending on performance and die size. What will differ it from Navi 11 is its size, performance, and which one was designed first, and which second.

And I will quote myself again, this time making the part that you missed more visible:



Do you want to make another round?

I will quote you, then:

This is the very definition of binning the same sillicon. What about that what you wrote before, about transistors, pipelines, size of cores, circuit: are there any differences between desktop and mobile Vega. Any differences between the Vega Pro 56 in iMac pro and the Vega 20?

Why you were talking about Vega 12 being cut down part of Vega 10 and differently binned, then, IF you knew they are different designs...?

Yes, there are. AMD claims that Vega Pro 16/20 comes with "optimised geometry engine" compared to the desktop Vega. The chip itself is the Vega Codename 12 that was leaked some time ago, while desktop 56/64 are Vega 10. All these numbers are rather confusing.

From what I see on Apple website, if says: "Enhanced Compute Engine", which pretty much refers to Rapid Packed Math.

 

[Ploki]

Kinda lame of Apple to not have put the Vega in the MBP 2018 release 2-3 months ago. Annoying for those of us who just bought the 2018 MBP.

It pisses me off beyond comprehension

 

[leman]

It pisses me off beyond comprehension

Well, it wasnt ready yet. Would you prefer them delaying the laptop release until December? Besides, it seems that the initial availability will be very low and we still dont know how much Vega option will cost...

 

[Howard2k]

Well, it wasnt ready yet. Would you prefer them delaying the laptop release until December? Besides, it seems that the initial availability will be very low and we still dont know how much Vega option will cost...

It would be nice if Apple would plan their MacBook release schedule around my specific personal buying patterns. I don't think that's too much to ask.

 

[ocnitsa]

Well, that's what the guy told me. Price difference between actual configurations. Maybe I misunderstood, not sure? I guess I will go to the Apple store in late November and ask them there.

There is no way that at "any time" one could do this...You couldn't buy a spec'ed MBP and then come back in a year and get a new one of the same year with a slight spec bump for a small amount of money that equated with the price difference. Likely, this refers to a reasonable amount of time between a purchase and a new feature being offered in which one can make an argument that I would have bought with said feature if I'd had the opportunity.

 

[koyoot]

From what I understand it was AMD who was not ready with release of those GPUs in June. Apple could hold off the update just for those GPUs, but decided not to.

Well, they could do it as Luis Rossmann says: cause your' still gonna' buy them!

 

[Ploki]

Well, it wasnt ready yet. Would you prefer them delaying the laptop release until December? Besides, it seems that the initial availability will be very low and we still dont know how much Vega option will cost...

Yeah i would, and i dont really care - i paid 5,3k € for it. 400€ wouldn't have made much difference at that point for either vega option.
And the point is if i was completely satisfied with the computer i wouldnt care - the display flickers (clean install not even icloud sign in), audio drops out in logic on external and internal card on same projects where 2012 worked fine, and i missed the 14 day return.
i had it for a month.
okay, so Logic doesnt get any more tracks with 2,9 than with 2,2 - i can live with that. It never hits 4,8 GHz, okay fine.
But damn, having it turned "old" one month after purchase and before its working properly? Gee, thanks.
They released a new laptop before i even got a replacement for a faulty magic mouse i bought with the first one.

I shouldn't have to jump through hoops to prove the laptop doesnt work as expected - its 5,3k professional machine, display shouldn't flicker and if it flickers on a clean install you better issue me a replacement yesterday.

I can buy machines that don't work properly for less than 5 grand (and even then I'll probably get a replacement faster).

It's been less than 4 months since they released the 2018 MacBook pros... They could release them with a statement "coming with vega in november".
Not have customers buy new machines just so they can buy one without a 2y old GPU few months later. And that's for those that got them on the date. For every BTO that's 2 months after.

If i were at least 95% satisfied with the laptop and didnt spend last three weeks on the phone with apple support being asked "are you using an external calibration device" 5 times *after* i sent them a video+screen recording of the problem with email explicitly stating "only stock system witihout 3rd party hardware was used" i'd feel bad about it, but i dont.
i bought a 5k machine that doesnt work as it supposed to and its been superseded before iti even started to work properly. I'm beyond pissed
[doublepost=1541008502][/doublepost]

Well, that's what the guy told me. Price difference between actual configurations. Maybe I misunderstood, not sure? I guess I will go to the Apple store in late November and ask them there.

wait what?
so i can get vega chip upgrade on my laptop?

 

[Thysanoptera]

Why you were talking about Vega 12 being cut down part of Vega 10 and differently binned, then, IF you knew they are different designs...?

I never said it is binned, you must have confused me with somebody else. I said your description of differences makes it look like a definition of a binned part, implying that there are no architectural or feature advancements that would explain its higher performance/watt. To me Vega 56 to Vega 20 looks like rx580 to rx560 and I’m eager to learn if this is in fact true. And yes, before you start, they were not binned either. Just leave binning out of it.

 

[MBX]

Not too much sad about it though, tbh. The base model GPU is more than capable. I'm even running Far Cry 5 (via Bootcamp) at Ultra Settings on everything and no frame drops at all.

 

[koyoot]

Just for those who are interested.

Vega 10 was GFX 900. Vega 15FF(Fenghuang Raven, engineering Platform) was GFX 902, Vega 12 is GFX 904, and Vega 20(HPC part) is GFX 906.

 

[stevemiller]

Well, on a long shot I gave apple support a friendly call to see if they could make any exceptions on my September purchase, but it was a little too long after the 15-days. Sigh. I guess I'll go back to my original plan of saving up for an egpu.

At least is it a safe bet that if I grab a Vega 56 or 64 at some point it'll ultimately pack more "oomph" than this mobile Vega 20? (just trying to put a positive spin on the disappointment)