Read more.Quote:
YouTuber reckons the laptop APU will feature 4x Zen 2 cores and up to 10x RDNA 2 CUs.
Printable View
Read more.Quote:
YouTuber reckons the laptop APU will feature 4x Zen 2 cores and up to 10x RDNA 2 CUs.
These rumours look wrong to me.
So the are supposed to have done the work for RDNA2 but stuck to Zen 2 cores?
Surely Renoir and removing 4 Zen2 cores would have made far more sense.
Some people suspect they're based on the PlayStation and Xbox Soc's, at least that's what was reported last month based on an AMD's engineers posting a Linux boot log.Quote:
Originally Posted by kompukare
What's wrong with Zen 2? It's plenty fast enough and if they can manufacture a cheap, low power variant and use it in a mainstream laptop, it'll still be better than whatever Intel can do with their current tech.Quote:
Originally Posted by kompukare
I'm sure the dual core Athlon 3000 series die started off as an embedded part for controlling advertising displays etc. That might explain the oddball look of these.
Would be nice if all under-yielding console parts didn't go to landfill.Quote:
Originally Posted by Corky34
Sony are now boasting they sold 7.6 million up to end of Q1 (2021-03-31), and expecting to sell 14.8 million over the next full year.
As they were meant to have had really poor yields compared to Microsoft with the bigger die (Sony pushed the clocks at the last minute after they found out Microsoft had gone for CUs), there must be millions of chips already which could be binned rather than binned.
In the Sony mix there could be anything from fully working dies which couldn't hit Sony's clocks, to variants like 6 to 4 cores with various amounts of functional CUs.
Maybe because of product segmentation or lead time, but a Zen2 CCD is around 72mm² while the Zen3 CDD is around 84mm². So half of that is a difference of 4mm². Which is not going to make any Zen2 based APU much cheaper to make.Quote:
Originally Posted by bae85
IIRC, the video suggests they will be using one CCX, not one CCD. That's the size difference.Quote:
Originally Posted by kompukare
Might be a business decision mainly... who knows.Quote:
Originally Posted by kompukare
Of course, all their APUs have always been monolithic.Quote:
Originally Posted by loftie
The comparison of Zen 2 versus Zen 3 was to guess how much space they're saving by going with the Zen 2 cores.
BTW, the Renoir versus Cezanne is 156mm² and 175mm², which would imply that Zen3 would be 10mm² extra for four cores, Cezanne doubles the L3 cache from 8MB to 16MB which makes it harder to calculate the CPU size changes.
Why would they sue the FF3 socket?
I rather enjoyed that.Quote:
Originally Posted by kompukare
As for Zen 2 and 10 CUs, I'd buy one in a laptop if the price was right. I think Zen 2 is plenty for most uses you'll see in a laptop. For high end stuff I'd want Zen 3 absolutely as you're paying a fair bit and you get the rewards towards the end of the useful life. The biggest issue I'd have here is 4 cores and whether it's enough but again - price is king. I currently have two cores on my laptop. Four is moar better.
:)Quote:
Originally Posted by philehidiot
Yes, was considering highlighting that somewhat with [sic] or something, but decided against it.
Yes, obviously if the price is right Zen 2 is still a great performer.
Just hope the reason wasn't segmentation as that is a very dangerous road to go down on (I would argue that Intel's obsession led to them turning down Apple for the iPhone manufacturing, caused Atom to be so neglected that even thrown billions away on contra revenue could revive it, and saw Intel overtaken by TSMC).
It is as likely just the current semiconductor industry mess getting AMD to just ship what they can, and this is something they had.
Segmentation doesn't make that much sense atm. I did get the impression from somewhere that the Zen3 APUs went really well, if this didn't and was late then the rate new parts are coming out...
I'm struggling with the idea that this is come console cast off though. Laptops are a pain to design, and benefit from standard pinouts. I can't see a console part fitting a standard laptop motherboard, so who would use it?
I don't think it's a cast off as that TPU article says it's based on console chips so i assume they tweaked that design and spun it into a new chip.Quote:
Originally Posted by DanceswithUnix
That doesn't really make any sense though.Quote:
Originally Posted by Corky34
AMD made the games console SoCs by taking GPU and CPU elements from their currently library of designs, and customising them.
Why on earth would you re-customise something? By the time you have undone some of the custom bits, you could have just started off back at the company library of IP blocks and started putting things together from there onto a clean sheet which is far less error prone.
I would imagine for the same reason Intel have been re-customising the Core design for the last decade or so. I'm not going to pretend to know their reasons but a Van Gogh chip using RDNA 2 is defiantly a thing.Quote:
Originally Posted by DanceswithUnix
OK, I'm not disputing it using RDNA 2, just that I don't see a causal link to console chips.Quote:
Originally Posted by Corky34
Let me put it another way. SoC design is hierarchical, and an HDL allows you to create instances of hardware blocks in your design. So in an xbox chip the top level design will call up 2x Zen2 CCXs and a bunch of RDNA CUs. In this Van Gogh case, you get one CCX and something like 10 RDNA2 CUs. My point is that there is an absolute ton of other stuff going on in an SoC in terms of PCIe connectivity, memory interfacing, SerDes lanes for SATA + USB + etc.
So, if I wanted to make a low cost laptop chip, I would start with AMD's usual fabric, hang the usual laptop IO and memory off it connected to the usual AMD laptop pinout (to make integration and driver issues easier) and create instances of the required quantities of CPU and GPU cores. If it just happened to end up sharing some blocks with a games console, then that would be entirely by accident.
Now perhaps this isn't an accident, perhaps Microsoft want to make an Xbox compatible Surface. But more likely, AMD saw some risk in using Zen3 in a budget market where Zen 2 is plenty, but RDNA2 is more cache friendly and would be ideal in an APU with limited memory bandwidth.
I get the feeling people have taken me saying they're based on the PlayStation and Xbox Soc's when the TPU article said "The APU is believed to be a design similar to the one found in the SoC of the latest PlayStation 5 and Xbox Series X/S consoles" a little too literally, i did provide a link to the article and perhaps should've quoted that line verbatim rather than simply saying "based on", i expected people would read the article rather than take my word as gospel, apologises.Quote:
Originally Posted by DanceswithUnix
I assumed "a design similar", "based on", the console SOC's would be interpreted in the same way as i did, that the APU having more in common with the features of console SOC's than previous gen APU's.
That was pretty much my thought - it's likely to essentially be an embedded/industrial part that might find its way into the low-end consumer stack. It would also explain using a more established core architecture to allow more mature drivers/compiler support etc.Quote:
Originally Posted by DanceswithUnix
I suspect you're right about RDNA 2 being used as it's low-bandwidth-friendly. Also given the core clocks we've seen it hitting in dGPUs I wonder if it can hit 1~1.5GHz with practically no voltage/power draw, allowing them to really can power draw across the SoC: I've seen comments around 7.5W - 15W target, but I'd hope 4 Cores + 10 RDNA CUs could go lower than that...
They've never been wrong in the past :rolleyes:
Lots of noise from that guy, very little signal. I'd like to buy a cheap 4C zen2 APU, and it's probably a lot easier to put a zen2 CCX in a design than half a zen3 CCX, but why would AMD waste 7nm wafers on this in the current climate?