Det0x
Golden Member
- Sep 11, 2014
- 1,232
- 3,889
- 136
I think and hope AMD brings out a Zen 3+ on 5nm as the final hurrah for AM4.
5nm Zen3+ on AM5 seems more likely.
If AMD is going to rehash Zen3 on N5P (or similar; AMD may get their own custom node again), then it'll be AM5.
Nice. 5950x also wins for the overall score vs Rocket Lake, yet Rocket Lake has a slightly better ST score. Anyone want to guess which way Userbench will adjust their scores?
What a huge improvement over the 3950x in the latency ladder.
Warhol seems to fit thet role, though it's still somewhat unclear what it entails. On top of the new I/O die I also would have thought they'd want to use a 6nm versions of Zen 3 Compute die on at least some higher-end SKUs. (as the masks are compatible it should be a minimal effort) but it looks like it's not the case. Perhaps 6nm isn't ready by that time or is too expensive.
These are completely delusional. Zen 3 is N7 design. Also cost, capacity...
Could be N6 though, which is design compatible, but I wouldn't bet on it.
Hmm, yeah it seems that TSMC indeed hasn't made any performance claims about N6.N6 is unlikely. It isn't any more performant than N7, it's less performant tnan N7+ and N7P, and there's no way for us to know how much better AMD's custom N7 variant is than any of those three. But N6 would probably be a step backwards.
At the very least, the lower power-consumption will help clock the chips higher (all-core turbo, not single-core). I also find it hard to believe that trading that 15-20% higher density for performance (e.g. keeping the die-size) would result in zero performance gain.Last year, TSMC announced the N6 node. This is an EUV-based node, but unlike N7+ which is entering its 2nd year of production, N6 is designed to be the easy migration path from N7 as its design rules are fully compatible with N7. In fact, TSMC considers N6 to be part of the “N7 family”. N6 is said to provide around 15-20% higher density with improved power consumption compared to N7, albeit no iso-power or speed comparisons where provided. We believe the density gains are obtained from switching from double diffusion breaks to a single isolation line. TSMC did note that N6 has 1 more EUV layer compared to N7+.
In comments made during the quarterly call, CC Wei, TSMC's CEO and vice chairman noted that “most of the customers in the N7 will move to N6.” In fact it sounds like TSMC's N6 node is set to become another one of TSMC's popular, high volume nodes, with Wei further stating that “from that day on probably, the N6 will pick up all the momentum and pick up all the volume production.”
This is likely due to a minor BKM or PDK update that allows TSMC/AMD to tune the process for a better voltage/frequency curve and bin a single CPU slightly higher.
Nice OC on this 5600X.
At the very least, the lower power-consumption will help clock the chips higher (all-core turbo, not single-core). I also find it hard to believe that trading that 15-20% higher density for performance (e.g. keeping the die-size) would result in zero performance gain.
About the Custom AMD 7nm. What did you mean by that?
Finally some news! (but without any news)
Robert Hallock drew the Zen 2 CCD completely wrong in that video, and I don't trust the Zen 3 layout either as it doesn't look like the one slide we saw. It's something he should know like the back of his hand. This one mistake is grievous enough to give me the opinion he should probably be replaced.
Trying to give an accurate representation of both while still differentiating the 2 architectures would probably just confuse the intended audience.
Any bets how long it takes userbenchmark to add tweaks specific to Ryzen 5000?
Is the intended audience children with Down's syndrome?
Anyone watching that video is at least a somewhat curious. It's either an inexplicable screw up, or a total misunderstanding and underappreciation of the audience.
It's in AMD's interest to build curiosity and interest about tech. A more informed populace buys more AMD products in general.
Is the intended audience children with Down's syndrome?
Anyone watching that video is at least a somewhat curious. It's either an inexplicable screw up, or a total misunderstanding and underappreciation of the audience.
It's in AMD's interest to build curiosity and interest about tech. A more informed populace buys more AMD products in general.
That's 4 percentage points lower compared to the tweeted screenshot but this 5600X is still leading the ranking. They need to work harder on the tweaks! (What will happen when the higher end Ryzen 5000 chips get added though? )
Is the intended audience children with Down's syndrome?
Anyone watching that video is at least a somewhat curious. It's either an inexplicable screw up, or a total misunderstanding and underappreciation of the audience.
It's in AMD's interest to build curiosity and interest about tech. A more informed populace buys more AMD products in general.
https://hwbot.org/submission/4591800_jumper118_cinebench___r15_ryzen_5_5600x_2040_cbNice OC on this 5600X.
More info from the person that posted this on HWBot.Yes 4.6ghz at 1.2v in bios drops down to 1.16v under load. 1.25v in bios to get 4.7ghz stable. Only crash in X265 4k on overkill, so I lowered it to 4.65ghz then it was fine.
https://hwbot.org/submission/4591800_jumper118_cinebench___r15_ryzen_5_5600x_2040_cb
More info from the person that posted this on HWBot.
why did we invent a new word form when "it doesn't offer any more performance than N7, it's lower performing than N7+" would suffice?N6 is unlikely. It isn't any more performant than N7, it's less performant tnan N7+ and N7P, and there's no way for us to know how much better AMD's custom N7 variant is than any of those three. But N6 would probably be a step backwards.
See above, N6 would likely bring about clockspeed regressions and other ugly side-effects. There's no guarantee that there will be a Zen3+ ever. But eventually they're going to have to move to 5nm.
https://hwbot.org/submission/4591800_jumper118_cinebench___r15_ryzen_5_5600x_2040_cb
More info from the person that posted this on HWBot.