Recent content by Jan Olšan

You should recheck, X925 was definitely advertised as 6x128 at launch, other info (Zen 5) also looks wrong. Oryon doesn't support SVE.

Note that these numbers don't nearly tell the whole story. Usually the pipes or units are differentiated. For example all can do 1 SIMD integer add per cycle, but only some can do integer multiplication. floating point ops quite often aren't available on all pipes. Shuffle ops generally have...

I could spitball 10% then, would you also be content with that? :) Not sure those official Geekbench results are well chosen anyway, since the scores usually have spread and kinda feels the m4 result is closer to the top of the spread and the 9950X to the lower end. That's the general problem...

I would love free 1950X. It's a historically significant part.

If there is industry worth saving by by bailouts and public spending support, this is it, though. All those "stargate" idiocies that they boast about are pretty stupid investments, if they are willing to let Intel go the way of GloFo at the same time, because somehow everyone throws around...

Where? official charts say 3385 or 3400 depending on which score you look versus 3912

Is that so? I'm feeling somewhat skeptical...

How? Metal is a NIH Apple product, making it an API that game devs have to support on top of the mainstream APIs that already cover most of the market. What about that makes targeting the API better environment to develop in? What do you base this on? Do you think Apple has more mature tools...

Is there any proof they aren't made from the same masks as desktop ones?

It has nothing to do with being fanless but with lowering the current loads put on the battery, so it's the lower TDP and lower PL2s/PPTs help. High-wattage processors and dGPUs do harm. The higher temperature associated with loading a fanless device may actually harm the battery's longevity.

But what if it is 160MB dual-stack cache dies under one CCD and still plain 32MB CCD for the other 8 cores? I doubt I was the only one who got the idea. You may not like it but this may be what peak performance looks like. You would not see a performance degradation in the vast majority of...

Very much agreed, it's just a tool, calculating that by dropping SMT you can make a core smaller and stuff more of them on chip to compensate, that is likely a very valid approach If you mean that there will be a mix of P and E on a server chip or worse, that instead of SMT, you would basically...

Why do people keep putting out these opinions without actually knowing if the business was profitable? If it was, there is no reason to just dump it because it is not cool and muh small lean company. It may be something they are forced to shed to get cash for other stuff but it harms the...

Likely just said the wrong term despite thinking the right concept. You are likely quite under pressure during these calls and presentations. That 55% share is likely correct, or at least, correct wrt available data. Gotta remember that the data we see from Mercury Research are just for AMD...

Panther Cove then, or whatever is the codename for Diamond Rapids. It's not the same as Lion but given how Intel ticks and tocks work it should be fundamentally the same core. Lion Cove was said to have optional HT last year, clearly for servers.