Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Intel 8th Gen Core with Radeon RX Vega GPU Presentation

videocardz.com

Post reply on HN
Print view View original

21–29 of 29 posts

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#21

Post by monocasa »

bryanlarsen wrote: post #19
monocasa wrote: post #6

Only 8x PCI-E 3.0? That's a little on the low side for a regular card, not to mention chips on the same interposer, no?

AFAICT, Coffee Lake has 24 PCI-E lanes, 8 of which are used to connect to the chipset. If they used all remaining 16 for the GPU, then NVMe and thunderbolt connections would have to go through the south bridge, affecting latency and throughput.

I guess I would have preferred to see it be a client on QPI and be cache coherent with the rest of the system, rather than using PCI at all, but yeah, that's asking a bit much on this super short timeframe.

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#22

Post by danmaz74 »

monocasa wrote: post #20
danmaz74 wrote: post #15

Earlier quoted context omitted.

Isn't that the kind of optimizations that you can only make for consoles, where you know that all users have the same hardware?

No, modern GPU APIs give you plenty of control over data layout, and it's the model AMD had been pimping for a while with their 'HSA' push.

I wasn't thinking about technical constraints, but economic ones. The market that would be able to benefit from this kind of optimization would be very limited as a % of PC gamers (which is already a limited market compared to consoles AFAIK).

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#23

Post by fleetfox »

monocasa wrote: post #13
fleetfox wrote: post #9

Earlier quoted context omitted.

It's enough. Even top end discrete cards can't saturate 16x PCI-E 3

I guess I'd like to see much tighter integration between the two chips so that we don't have to write code that avoids the PCI bus at all costs. Today you fit your entire working set into VRAM and don't touch it from the CPU, so it's sort of hard to say "it's not a deal we don't saturate the bus anyway" when that's only because we bend over backwards to not touch the bus in the first place.

Could you elaborate what you mean by tighter? Isn't this the whole point MCM designs?

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#24

Post by dogma1138 »

neltnerb wrote: post #16
monocasa wrote: post #13

Earlier quoted context omitted.

I guess I'd like to see much tighter integration between the two chips so that we don't have to write code that avoids the PCI bus at all costs. Today you fit your entire working set into VRAM and don't touch it from the CPU, so it's sort of hard to say "it's not a deal we don't saturate the bus anyway" when that's only because we bend over backwards to not touch the bus in the first place.

I would imagine that they took the best GPU chip they could afford the space and power for, and found that 8 channels was enough for 99% of their customers? I assume adding more channels takes space, perhaps the GPU would have had to be lower performance to add more channels? (speculating)

It's not like they had an option, NVIDIA doesn't do semi-customs you either buy their SOCs or their GPUs as a package.

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#25

Post by dogma1138 »

bryanlarsen wrote: post #19
monocasa wrote: post #6

Only 8x PCI-E 3.0? That's a little on the low side for a regular card, not to mention chips on the same interposer, no?

AFAICT, Coffee Lake has 24 PCI-E lanes, 8 of which are used to connect to the chipset. If they used all remaining 16 for the GPU, then NVMe and thunderbolt connections would have to go through the south bridge, affecting latency and throughput.

Intel doesn't connect to the chipset with PCIe, it does that with DMI.

Coffee Lake only has 16 PCIE lanes on the CPU, 4 DMI Lanes (about equivalent to 4 PCIE 3.0 lanes) which connect to the PCH and 24 PCIE 3.0 lanes coming from the chipset.

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#26

Post by dogma1138 »

lettergram wrote: post #8

I mean.. I wouldn't trust those benchmarks at all. It does seem like good news for AMD though!

I'm not sure how good it is for AMD since their Ryzen G series APUs which also aim for the 45-60W target come with a much weaker GPU (11 CUs and no onboard memory).

Sure they would be cheaper but it doesn't really matter that much the OEMs price the laptops too close together now for that to make a difference.

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#27

Post by bryanlarsen »

dogma1138 wrote: post #25
bryanlarsen wrote: post #19

Earlier quoted context omitted.

AFAICT, Coffee Lake has 24 PCI-E lanes, 8 of which are used to connect to the chipset. If they used all remaining 16 for the GPU, then NVMe and thunderbolt connections would have to go through the south bridge, affecting latency and throughput.

Intel doesn't connect to the chipset with PCIe, it does that with DMI. Coffee Lake only has 16 PCIE lanes on the CPU, 4 DMI Lanes (about equivalent to 4 PCIE 3.0 lanes) which connect to the PCH and 24 PCIE 3.0 lanes coming from the chipset.

I over simplified. DMI is a PCI Express variant. I thought it had increased to 8 lanes in Coffee Lake, but I was mistaken. That makes it even more important to save some of the CPU PCIe lanes for NVMe and thunderbolt.

Re: Intel 8th Gen Core with Radeon RX Vega GPU Presentation

#29

Post by monocasa »

fleetfox wrote: post #23
monocasa wrote: post #13

Earlier quoted context omitted.

I guess I'd like to see much tighter integration between the two chips so that we don't have to write code that avoids the PCI bus at all costs. Today you fit your entire working set into VRAM and don't touch it from the CPU, so it's sort of hard to say "it's not a deal we don't saturate the bus anyway" when that's only because we bend over backwards to not touch the bus in the first place.

Could you elaborate what you mean by tighter? Isn't this the whole point MCM designs?

Right, exactly, but this chip is still as far away from a software perspective as a discrete card would be.
Post reply on HN