Were Ryzen 8000 CPUs a big mistake from AMD?

[Nhonho]

Why didn't AMD put at least 32 MB of L3 cache on them?

With only 16 MB of L3, the Ryzen 8000 CPUs are slow.

Of course, with 32 MB of L3 the chip would be bigger and more expensive to manufacture, but it would performance better and people would buy it anyway and would not buy CPUs from competitors.

One thing is for sure, it is not by making bad savings on its products that AMD will be able to sell many CPUs and gain more market share.

 

[eidairaman1]

More griping, this is just speculation, wait until they are in hands of the people.

 

[Lew Zealand]

They're not a mistake, instead they're a targeted product.

The R 8000s are laptop-sourced chips like the 5x00G, and earlier G-suffixed CPUs. Monolithic, no IO Die, no separate CCD die, based on the Phoenix 7xxx series laptop APUs. They easily could have been named the 7700G and 7600G like the 5xxxg series but AMD just went with a separate number for naming, probably a better idea than the mess that 5xxx and 4xxx naming is.

They're great for what they are as you'll be building a USFF PC from them with no dGPU. Turns out I'm using a Phoenix-core PC right now, but one with the 7840HS in it, which is almost exactly what the 8700G is (a 7940HS).

 

[dirtyferret]

I'm going to ask this in all seriousness, are you new to AMD's APUs? If you are, these are niche CPUs to fill a specific sector of the market and they really have no "competitors". If not, than the question really is self explanatory.