News Intel: Xe 2 Family Includes Xe-LPG and Xe-HPG GPUs With Slight Variations

[Admin]

Intel vows to reduce IP variations between different GPU microarchitectures.

Intel: Xe 2 Family Includes Xe-LPG and Xe-HPG GPUs With Slight Variations : Read more

 

[evdjj3j]

"There is a Xe and there is a Xe 2 and in that Xe 2 generation there is a Xe-LPG and there is a HPG (…) and there a slight variations (…) which is our big learning,"

That would have probably made a lot more sense if you wouldn't have left the .... out.

 

[bit_user]

"There is a Xe and there is a Xe 2 and in that Xe 2 generation there is a Xe-LPG and there is a HPG (…) and there a slight variations (…) which is our big learning,"

That would have probably made a lot more sense if you wouldn't have left the .... out.

When you're quoting someone and you want to omit parts of it, you need to indicate where you edited their statement. That's standard journalistic practice.

The reason for such edits is to clarify or bring out the essence of a particular statement. So, you're basically affirming that.

Update: upon visiting the original HardwareLuxx article to get more of the context of these quotes, I noticed the "(...)" are actually present in their article.

 

[cyrusfox]

Another day, another article suggesting Intel dGPU business is safe for now, I hope MLID got this wrong, demise/exit of Intel discrete graphics.

As battlemage is so far along, maybe it would be more likely for Celestial or Druid for the exit point though. Also what if dGPU does disappear for the mid to low end thanks to direct die attach of future GPUs (Meteor Lake)?

Glad to see price drop on A750 as well as driver work maturing. I hope they keep all the fixed funciton accelerators on the small gpu die for BattleMage and I am very interested in the rumored half stepping ACM-G12 as the die is too large for ACM-10 to justify a product @ the $250 mark, Good for Intel to clear inventory but I would imagine the Margin is 0 or negative on these products for them.

 

[bit_user]

Tom Peterson: "... we would have been better off have we rigorously said 'you know what, we gonna give up something', like give up some differentiation in the high end or we are going to have some overhead in the low end. We are gonna just have one thing and it goes everywhere unmodified."​

​

Um, good luck with that. Both AMD and Nvidia have decided to have one microarchitecture for datacenter and a separate one for consumer. Granted, they don't have the three variations he mentioned, but I think 2 makes sense.

One reason it's probably wise to unify iGPU and consumer dGPU is that it lets game developers target only one microarchitecture for the brand/generation.

 

[bit_user]

Glad to see price drop on A750 as well as driver work maturing. ... I am very interested in the rumored half stepping ACM-G12 as the die is too large for ACM-10 to justify a product @ the $250 mark,

I am very interested in an Alchemist refresh. I hope it doesn't get canceled.

 

[InvalidError]

Glad to see price drop on A750 as well as driver work maturing. I hope they keep all the fixed funciton accelerators on the small gpu die for BattleMage and I am very interested in the rumored half stepping ACM-G12 as the die is too large for ACM-10 to justify a product @ the $250 mark, Good for Intel to clear inventory but I would imagine the Margin is 0 or negative on these products for them.

The A750 is the outlet for A770 dies that don't make the grade, can't really have a separate die for it. What would make more sense for a refresh lineup would be a die that more cost effectively covers the A380-580 gap.

IMO, DG2-128 aimed too low to be viable beyond basic desktop graphics. Intel really needed a DG2-256 to turn some heads at the $200-ish mark, knocking the RX6500 and anything weaker than that at the time out of the park.

 

[bit_user]

What would make more sense for a refresh lineup would be a die that more cost effectively covers the A380-580 gap.

Or, maybe there are chip bugs that are holding back performance. That's my hope.

 

[-Fran-]

Learning lessons is great. Let's hope that is actually true.

Now the important matter is going to be the "when". Given how badly they fumbled all of the ARC launch dates and the expectations, I really really hope they learned their lesson.

Regards.

 

[InvalidError]

Or, maybe there are chip bugs that are holding back performance. That's my hope.

I wouldn't mind a more palatable sub-$200 GPU based on a DG2-256 die free from whatever known hardware shortcomings may have been identified in "Alchemist-gen1" assuming it becomes available before I decide (or am forced) to retire my GTX1050.

 

[usertests]

Another day, another article suggesting Intel dGPU business is safe for now, I hope MLID got this wrong, demise/exit of Intel discrete graphics.

What he has been saying lately (or from the beginning idk) is that there will be a low-end Battlemage die, not multiple, not competing with high-end. And Intel will obviously never cancel integrated graphics. That's consistent with Intel's PR.

It would be nice to see more competition under $200. The Arc A380 is already an interesting alternative to the RX 6400 and 6500 XT, with more VRAM, more PCIe lanes, more video support, but worse performance. If AMD makes a 7500 XT with 6 GB, PCIe 4.0 x8, and the missing decode/encode, there would be nothing to complain about except the prices.

 

[bit_user]

What he has been saying lately (or from the beginning idk) is that there will be a low-end Battlemage die, not multiple,

That's not what microarchitecture means. It's referring to the design of the individual Xe cores, for instance. This is consistent with his focus on IP reuse, as opposed to reuse of the final silicon dies.

If you read the original German article, you can see his full statements in English, and that's all he really said. And he listed only 3 different variations on Xe cores.

The Tom's article is partly to blame for some confusion, with how it's talking about "nine different GPU variations".

 

[-Fran-]

Probably related to the topic:

View: https://www.youtube.com/watch?v=b-6sHUNBxVg


Looks like Intel has been doing a good job with the drivers and they're keeping (or trying to keep) cadence. Still not perfect and not a magic silver bullet to fix everything (performance and usability), but still better than the day-1 mess. This gives me a bit more hope for Battlemage and the potential Alchemist refresh cards rumored to be coming next.

Regards.

 

[cyrusfox]

What he has been saying lately (or from the beginning idk) is that there will be a low-end Battlemage die, not multiple, not competing with high-end. And Intel will obviously never cancel integrated graphics. That's consistent with Intel's PR.

That is not what he said at the beginning, yes he has softened his initial "cancellation" [backpedaled the extreme claims], but his original words can be heard and seen Here where he clarified his singular stance, he was the first to claim death of Arc at the time:

View: https://www.youtube.com/watch?v=cZr_LWAlDkg

"Datacenter stays for now, but the dedicated gaming card line may be cancelled before Celestial even gets a chance."
These are other claims he makes in the video (not exhaustive):

• Claims no more discrete consumer card [multiple times throughout video, few snippets below]
  • Arc is Finished [4:08]
  • Intel is ending Arc Discrete [4:28]
  • Arc is not long for this world anymore [7:31] - At least in a way that would make any excited...
  • The markets not working for them(Excess card inventory), they don't believe they can get it(Arc) working, and its (Arc) done. [11:14]
• No Battlemage multiple die lineup[2:07]
  • Battlemage is going to be a single die (either low-end or midrange) [3:10]
• Intel can't get multi-die working [3:41]
  • Multi die for enthusiast isn't going to happen [7:14]
• Arc Alchemist is very low volume [6:30]

Was he overly exaggerating his claims to ensure more clicks and capture more subscribers? This initial stance reveals his accuracy (or that of his sources) as we see ARCs fate unfold in real time. I would argue he is likely right in the direction (Going from 4 different GPU designs/dies design down to less appears to be true, just not a single design so far...) He made extreme claims, even though he has since tempered/ back pedal those statements, ARC is not dead yet.

It's great to see Arc being developed (drivers and new silicon), each piece of news that suggest that Arc is still alive and may even get feet to start running. But it is Intel, it could be killed, look at Optane and the other items Intel is killing (RISC).

 

[bit_user]

I would argue he is likely right in the direction (Going from 4 different GPU designs/dies design down to less appears to be true,

I'd argue the real number is probably 3. I think HPG and HP were basically the same core, with the main difference being whether or not there's fp64 hardware. The Hot Chips presentation shows that part is optional. They probably decided fp64 wasn't a big requirement of most HP customers and therefore they could substitute the HPG chips for them.

But it is Intel, it could be killed, look at Optane and the other items Intel is killing (RISC).

Intel is not killing RISC-V. They just ended a SDK product they made for 3rd party RISC-V IP. That was just to help the ecosystem get started, and now there are other efforts which can partially take its place.

Importantly, Intel followed-through with their launch of Horse Creek, which is a big win for anyone looking for a serious RISC-V development platform (i.e. not simply for embedded applications).