Recent content by AnotherReader

AMD has the solution for this since RDNA2: dedicated last level cache for the GPU. Let's see if they actually equip Strix Halo with one.

I really think Intel missed a trick here by not courting others aggressively for EMIB. They could have taken advantage of the shortage of TSMC's COWOS and its technical limitations.

While I agree that the fastest of the rumoured RDNA 4 SKUs is unlikely to be faster than the 7900 XT, I don't think that RDNA 3 has no improvements compared to RDNA2. RDNA3 improves upon RDNA2, but the compiler isn't good enough to make decent use of its dual issue capabilities. There are...

Hasn't Apple included AI inference accelerators since the A11 which was launched over six years ago? As the M series of SOCs are in that lineage, they have NPUs too. So what's new here?

Approximations are good enough for many purposes and transcendental functions include sine, cosine, reciprocal, and square root. These are used in 3D graphics as well.

It's an inherent shortcoming of NAND just like DRAM's latency hasn't decreased in decades though in DRAM's case, it's related to the interface as well. Lower latency DRAM is possible, but it would cost area and would require changes to the way DRAM is accessed.

@AleksandarK I believe Intel 7 still uses DUV. Intel 4 is their first process to use EUV.

I think we're using different metrics. I'm talking about revenues. Market capitalization is unreliable and is a hopeful indicator of future performance. By market capitalization, I expect Nvidia has a good chance of becoming the most valuable company in the world this year.

Without a doubt, Nvidia is the leader in AI. However, the biggest limiters for these AI chips are the interposer or COWOS as TSMC calls it and the HBM. Nvidia is very rich, but they are still not in the ballpark of the true big players: Apple, Google, and Microsoft. Of these, Google is the...

End the ban on Huawei and other Chinese companies and TSMC wouldn't even need Nvidia. AMD is using a node slightly inferior to what Nvidia is using and as usual, their top end GPU die is smaller than Nvidia's top end. Given how well RDNA 3 scales with increasing CU count, a 120 CU RDNA 3 SKU...

There's also the association, right or wrong, of using Apple with snobbery. As for the bug, if it is indeed unfixable in hardware for older M series SOCs, then that would indicate a lack of updatable microcode in Apple's designs.

Unlike the SSD manufacturers, this should have no discernable effect for the end users.

Both of them need each other, but of the two, TSMC is more indispensable. Nvidia went with Samsung and only came out on top because AMD didn't build as big a chip as GA102. Navi 21, sans the 128 MB of LLC, would have been closer to AD104 in die size. One can imagine that a bigger Navi 21 would...

I get that, but we have had unattractive male characters so why not have ordinary looking females as well. Let's not forget that in the USA, women are estimated to be close to 50 percent of the PC gaming user base. They are also estimated to be over 40% of the PS5 customer base.

I don't understand the fuss around Aloy's face. Was there such outrage when Geralt's face changed dramatically from Witcher I to Witcher 2 and then again in Witcher 3? I am also looking forward to the review by @W1zzard.