Soldered on ram and GPU.
-
In systems where memory speed are mismatched, the system runs at the slowest module's speed. So literally making the soldered, faster memory slower. Why even have soldered memory at that point?
I'd assume the soldered memory to have a dedicated memory controller. There's also no hard requirement that a single controller can't drive different channels at different speeds. The only hard requirement is that one channel needs to run at one speed.
-
Ye the soldered ram is for sure making me doubt framework now.
Apparently AMD wasn't able to make socketed RAM work, timings aren't viable. So Framework has the choice of doing it this way or not doing it at all.
-
The cache hierarchy has flopped? People aren't using swap?
NUMA also hasn't flopped, it's just that most systems aren't dual socket. Different memory speeds for the same CPU is not ideal and you don't build a system like that but among upgraded systems that's not rare at all.
Yeah, the cache hierarchy is behaving kinda wonky lately. Many AI workloads (and that’s what’s driving development lately) are constrained by bandwidth, and cache will only help you with a part of that. Cache will help with repeated access, not as much with streaming access to datasets much larger than the cache (i.e. many current AI models).
Intel already tried selling CPUs with both on-package HBM and slotted DDR-RAM. No one wanted it, as the performance gains of the expensive HBM evaporated completely as soon as you touched memory out-of-package. (Assuming workloads bound by memory bandwidth, which currently dominate the compute market)
To get good performance out of that, you may need to explicitly code the memory transfers to enable prefetch (preferably asynchronous) from the slower memory into the faster, á la classic GPU programming. YMMW.
-
Couldn't you just treat the socketed ram like another layer of memory effectively meaning that L1-3 are on the CPU "L4" would be soldered RAM and then L5 would be extra socketed RAM? Alternatively couldn't you just treat it like really fast swap?
Wrote a longer reply to someone else, but briefly, yes, you are correct. Kinda.
Caches won’t help with bandwidth-bound compute (read: ”AI”) it the streamed dataset is significantly larger than the cache. A cache will only speed up repeated access to a limited set of data.
-
Couldn't you just treat the socketed ram like another layer of memory effectively meaning that L1-3 are on the CPU "L4" would be soldered RAM and then L5 would be extra socketed RAM? Alternatively couldn't you just treat it like really fast swap?
Could it work?
Yes, but it would require:
- A redesigned memory controller capable of tiering RAM.
- OS-level support for dynamically assigning memory usage based on speed.
- Applications/libraries optimized to take advantage of this tiering.
Right now, the easiest solution for fast, high-bandwidth RAM is just to solder all of it.
-
Apparently AMD wasn't able to make socketed RAM work, timings aren't viable. So Framework has the choice of doing it this way or not doing it at all.
In that case, not at all is the right choice until AMD can figure out that frankly brain dead easy thing.
-
Yeah, the cache hierarchy is behaving kinda wonky lately. Many AI workloads (and that’s what’s driving development lately) are constrained by bandwidth, and cache will only help you with a part of that. Cache will help with repeated access, not as much with streaming access to datasets much larger than the cache (i.e. many current AI models).
Intel already tried selling CPUs with both on-package HBM and slotted DDR-RAM. No one wanted it, as the performance gains of the expensive HBM evaporated completely as soon as you touched memory out-of-package. (Assuming workloads bound by memory bandwidth, which currently dominate the compute market)
To get good performance out of that, you may need to explicitly code the memory transfers to enable prefetch (preferably asynchronous) from the slower memory into the faster, á la classic GPU programming. YMMW.
I wasn't really thinking of HPC but my next gaming rig, TBH. The OS can move often accessed pages into faster RAM just as it can move busy threads to faster cores, gaining you some fps a second or two after alt-tabbing back to the game after messing around with firefox. If it wasn't for memory controllers generally driving channels all at the same speed that could already be a thing right now. It definitely already was a thing back in the days of swapping out to spinning platters.
Not sure about HBM in CPUs in general but with packaging advancement any in-package stuff is only going to become cheaper, HBM, pedestrian bandwidth, doesn't matter.
-
Signal integrity is a real issue with dimm modules. It's the same reason you don't see modular VRAM on GPUs. If the ram needs to behave like VRAM, it needs to run at VRAM speeds.
Then don't make it work like that. Desktop PCs are modular and Framework made a worse product in terms of modularity and repairability, the main sales of Framework. Just, like… wtf. This Framework product is cursed and shouldn't exist.
-
Then don't make it work like that. Desktop PCs are modular and Framework made a worse product in terms of modularity and repairability, the main sales of Framework. Just, like… wtf. This Framework product is cursed and shouldn't exist.
There’s little point in framework selling a conventional desktop.
I guess they could have made another laptop size with the the dev time, but… I dunno, this seems like a niche that needs to be filled.
-
In that case, not at all is the right choice until AMD can figure out that frankly brain dead easy thing.
"brain dead easy thing"... All you need is to just manage signal integrity of super fast speed ram to a super hungry state of the art soc that benefits from as fast of memory as it can get. Sounds easy af. /s
They said that it was possible, but they lost over half of the speed doing it, so it was not worth it. It would severely cripple performance of the SOC.
The only real complaint here is calling this a desktop, it's somewhere in between a NUC and a real desktop. But I guess it technically sits on a desk top, while also being an itx motherboard.
-
In that case, not at all is the right choice until AMD can figure out that frankly brain dead easy thing.
Oh yeah I'm sure you could've done it no problem
-
There’s little point in framework selling a conventional desktop.
I guess they could have made another laptop size with the the dev time, but… I dunno, this seems like a niche that needs to be filled.
This is where I'm at. The Framework guy was talking about how very few companies are using this AMD deal because the R&D to add it to existing models wasn't very viable, you really only have the Asus Z13 so I feel like being ahead of the game there will be a benefit in the long run as far as their relationship with AMD. Plus they're also doing a 12-in laptop now as well, so it's not like they committed all their resources to this.
-
Apparently AMD couldn’t make the signal integrity work out with socketed RAM. (source: LTT video with Framework CEO)
IMHO: Up until now, using soldered RAM was lazy and cheap bullshit. But I do think we are at the limit of what’s reasonable to do over socketed RAM. In high performance datacenter applications, socketed RAM is on it’s way out (see: MI300A, Grace-{Hopper,Blackwell},Xeon Max), with onboard memory gaining ground. I think we’ll see the same trend on consumer stuff as well. Requirements on memory bandwidth and latency are going up with recent trends like powerful integrated graphics and AI-slop, and socketed RAM simply won’t work.
It’s sad, but in a few generations I think only the lower end consumer CPUs will be possible to use with socketed RAM. I’m betting the high performance consumer CPUs will require not only soldered, but on-board RAM.
Finally, some Grace Hopper to make everyone happy: https://youtube.com/watch?v=gYqF6-h9Cvg
Sound like a downgrade to me I rather have more ram than having a soldered limited one. Especially for consumer stuff.
-
I wasn't really thinking of HPC but my next gaming rig, TBH. The OS can move often accessed pages into faster RAM just as it can move busy threads to faster cores, gaining you some fps a second or two after alt-tabbing back to the game after messing around with firefox. If it wasn't for memory controllers generally driving channels all at the same speed that could already be a thing right now. It definitely already was a thing back in the days of swapping out to spinning platters.
Not sure about HBM in CPUs in general but with packaging advancement any in-package stuff is only going to become cheaper, HBM, pedestrian bandwidth, doesn't matter.
The thing is, consumers didn’t push Nvidias stock sky high, AI did. Microsoft isn’t pushing anything sane to consumers, Microsoft is pushing AI. AMD, Intel, Nvidia and Qualcomm are all pushing AI to consumers. Additionally, on the graphics side of things, AMD is pushing APUs to consumers. They are all pushing things that require higher memory bandwidth.
Consumer will get ”trickle down silicon”, like it or not. Out of package memory will die. Maybe not with you next gaming rig, but maybe the one after that.
-
Soldered on ram and GPU. Strange for Framework.
Not strange at all.
They're a business that makes its money off of selling hype to morons.
-
Sound like a downgrade to me I rather have more ram than having a soldered limited one. Especially for consumer stuff.
Sounds like a load of bullshit to feed useful idiots.
-
Not strange at all.
They're a business that makes its money off of selling hype to morons.
Just buy a ThinkPad, if you’re thinking about buying a Framework…
-
Sound like a downgrade to me I rather have more ram than having a soldered limited one. Especially for consumer stuff.
Looking at my actual PCs built in the last 25 years or so, I tend to buy a lot of good spec ram up front and never touch it again. My desktop from 2011 has 16GB and the one from 2018 has 32GB. With both now running Linux, it still feels like plenty.
When I go to build my next system, if I could get a motherboard with 64 or 128GB soldered to it, AND it was like double the speed, I might go for that choice.
We just need to keep competition alive in that space to avoid the dumb price gouging you get with phones and Macs and stuff.
-
Apparently AMD couldn’t make the signal integrity work out with socketed RAM. (source: LTT video with Framework CEO)
IMHO: Up until now, using soldered RAM was lazy and cheap bullshit. But I do think we are at the limit of what’s reasonable to do over socketed RAM. In high performance datacenter applications, socketed RAM is on it’s way out (see: MI300A, Grace-{Hopper,Blackwell},Xeon Max), with onboard memory gaining ground. I think we’ll see the same trend on consumer stuff as well. Requirements on memory bandwidth and latency are going up with recent trends like powerful integrated graphics and AI-slop, and socketed RAM simply won’t work.
It’s sad, but in a few generations I think only the lower end consumer CPUs will be possible to use with socketed RAM. I’m betting the high performance consumer CPUs will require not only soldered, but on-board RAM.
Finally, some Grace Hopper to make everyone happy: https://youtube.com/watch?v=gYqF6-h9Cvg
There's even the next iteration already happening: Cerebras is maling wafer-scale chipa with integrated SRAM. If you want to have the highest memory-bandwith to your cpu core it has to lay exactly next to it ON the chip.
Ultimately RAM and processor will probably be indistinguishable with the human eye.
-
You are correct, I’m referring to on package. Need more coffee.
No, it's a super common terminology almost to the point where I wouldn't really even consider it outright wrong to describe it as a SoC. It's just that the blurred distinction between a single chip and multiple chiplets packaged together are almost impossible for an outsider to tell without really getting into the published spec sheets for a product (and sometimes may not even be known then).
It's just more technically precise to describe them as SiP, even if SoC functionally means something quite similar (and the language may evolve to the point where the terms are interchangeable in practice).
-
System shared this topic on
