• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Question Zen 6 Speculation Thread

Page 154 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
LightningZ71 said:
This is why I suggest that a single CCD Medusa Halo would be a good fit. It's a completed design, so there's no new R&D to pay for. Unused IP in the chip can be fused off, reducing licensing costs. So, pay the smaller integration cost up front and buy the rest on a volume discount.
Click to expand...
No one is using 75mm² of N2P silicon for CPUs just to run games at 30/60 FPS.
 
Multiquote is breaking again sooooo

@adroc_thurston

Allegedly, Nintendo gets the SoC/chipset for Switch 2 at a cost of ~$150/unit. How much of that is going to NV and how much is going in Samsung's pocket?

@LightningZ71 @Thunder 57

So far as I know, AMD isn't actually taking wafers for console parts. They produce a design to meet the needs of the client (MS/Sony) and then license it to said client at a fixed upfront cost plus additional royalties per unit sold. The client - let's say MS - takes wafers from TSMC to produce the design. It's not like AMD has to cut into their wafer allocation of N3P (or whatever) to fill Microsoft's order. MS has to bid on wafers like everyone else.

AMD takes on near-zero risk assuming the design isn't total crap. They get paid upfront, so they don't have to move units to cover design costs. Royalties are just icing on the cake. They can hire more people if they need them and crank out as many console designs as they like without costs being a real limiter or risk factor.

As far as using Medusa Halo for a console goes . . . probably not. Don't be fooled by the miniPCs with Strix Halo and assume a nextgen unit like this would be a good console competitor. It would have too much CPU grunt to be a good console.
 
PCIe 6.0 on AMD (Venice by the sound of it):

www.techpowerup.com

Keysight Enables AMD to Showcase Electrical PCI Express Compliance up to 64 GT/s

Keysight Technologies, Inc. announced that it helped AMD accelerate the testing of electrical compliance for PCI Express (PCIe) specification for pre-production AMD server CPUs. By providing early access to beta software, Keysight enabled AMD to develop and test a PCI Express technology-capable...
www.techpowerup.com www.techpowerup.com
 
DrMrLordX said:
As far as using Medusa Halo for a console goes . . . probably not. Don't be fooled by the miniPCs with Strix Halo and assume a nextgen unit like this would be a good console competitor. It would have too much CPU grunt to be a good console.
Click to expand...
If Switch 1/2 and the post Gamecube consoles show us nothing it's that HW spec isn't really all that it's cracked up to be when it comes to selling consoles.

It's the platform UX (controllers + UI + fan noise etc), game library and the brand that determines its ultimate success.

Medusa/Strix Halo + Windows as a console is absolutely DOA - no question.

But either with a dedicated gaming OS, polished (invisible) drivers, a decent selection of modern games and a handful of exclusives + a good brand behind it could be successful.
 
soresu said:
Medusa/Strix Halo + Windows as a console is absolutely DOA - no question.

But either with a dedicated gaming OS, polished (invisible) drivers, a decent selection of modern games and a handful of exclusives + a good brand behind it could be successful.
Click to expand...
Even with the right software, the BoM for what you get is just too high. There's no need to throw that powerful of a CPU at a console.
 
LightningZ71 said:
Cost includes R&D sunk costs. Spending massive amounts (series x SoC was estimated to costs hundreds of millions in 2018 money) to develop a bespoke design has to be folded back into the cost of each chip. They may move 7 million units a year on average, something that takes hundreds of millions to develop adds a nontrivial cost to each chip.

If the package meets their needs, they can afford higher per unit costs on the back end by not fronting development costs up front.
Click to expand...
I think there's some confusion on that whole topic.

We are talking consoles. "Real" consoles are a piece of mass market hardware that aims to sell 100+ million in a life cycle that usually spans 5+ years. Steam Deck is far far from that, PC handhelds are far far far from that.

With this kind of mass market hardware you don't want to use off the shelf hardware. You want to cost optimize the hardware upfront to have as little cost as possible during the whole life cycle. You want your hardware to bespoke or you are doing it wrong. Microsoft got burned by that with the original gen XBox which they couldn't stop manufacturing quickly enough for that very reason.

For AMD it's free money. They get all the money to do all that necessary upfront work. With it they can hire more staff for that if necessary.

Microsoft is doing it wrong if it is selling too few consoles for the upfront cost to be worth it. Games sell consoles. Microsoft bought plenty of popular gaming IPs they could look at making good/better/any use of before considering launching their next console gen.
 
511 said:
The CPU would be likely single N4P Zen5 die and the GPU can use N3C with advanced packing would be lot cheaper for AMD to cook and should be cheaper to produce.
Click to expand...
You do understand that the sum total of advanced packaging wafer starts per year is woefully insufficient to support consoles?
 
DrMrLordX said:
There's no need to throw that powerful of a CPU at a console.
Click to expand...
Still, at least a small amount of V-cache (even 32MB) would be far better than no V-cache due to the much higher latency of GDDR6/7 RAM used in consoles. The elimination or minimization of CPU idle wait times between accessing data from RAM will help to smooth things over.

Another approach is to use tiered RAM. 16GB could be high speed DDR5-6000 CL26 RAM and the other 16GB could be GDDR6/7. Best of both worlds (lowest latency for executable code as well as highest bandwidth for textures) and leave it up to the developer which part they want to designate as common unified RAM storage area that can be accessed by both the CPU and GPU. Alternatively, they could do the tiering at the firmware level so that the game has no idea what it is accessing and the driver decides whether to place the data in system RAM or GPU RAM.
 
igor_kavinski said:
the game has no idea what it is accessing and the driver decides whether to place the data in system RAM or GPU RAM.
Click to expand...
Not good idea for consoles, splitting memory however gonna be necessary - getting VRAM to 24 GB might be cheap enough, but going much further not, hope they'll add at least 32-48 GB DDR5.
 
I kinda think that Zen 6 for nextgen console is a bit too optimistic. At best we will see Zen 5 or maybe they just repurpose Strix Halo with beefier GPU.

Zen 5 CPU 24 threads
RDNA4.5 GPU
Quad channel 32GB LPDDR6-9600
 
511 said:
they can always pack at Intel
Click to expand...
Their 2.5D start capacity is even more of a joke.
511 said:
that they stopped the expansion.
Click to expand...
that's because they be broooooooke as hell.
igor_kavinski said:
I kinda think that Zen 6 for nextgen console is a bit too optimistic
Click to expand...
It's H1'26 IP.
igor_kavinski said:
At best we will see Zen 5 or maybe they just repurpose Strix Halo.
Click to expand...
how would they repurpose a super high-end mobile part?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top