- Mar 3, 2017
- 1,747
- 6,598
- 136
Are you forgetting that LNC is a two node jump from GLC while Zen 5 is a half-node (at best) jump from Zen 4?
Node wise, Intel has the bigger jump,
Skymont yes, but as I said before...And where do you get the idea that Intel didnt "try" for a massive core arch refresh? Isnt that what both Lion Cove and Skymont are?
And tbh, some parts of LNC don't seem to be buffed as much as a usual Intel tock is, like the ROB capacity or the uop cache capacity
What makes you think we won't see a "massive" jump in perf/watt with ARL?If Arrow Lake doesnt show a massive jump in perf or perf/watt with a 2 node jump, its going to be a way bigger L than these results with Zen 5 we've seen thus far.
That was in decent part due to the regressed core counts. Also compounded by the facts that SNC itself wasn't a great arch, and that the arch was way too wide for 14nm, meaning vs SKL, u only saw perf/watt improvements at a relatively high upper end of the perf/watt curve of RKL.Last time Intel did major arch change on same node was Rocket Lake, and that was objectively their worst new CPU generation as far back as I remember.
12 Zen 5+ Zen5C cores don't have the same perf as 8 Zen 4 cores.Looks like the forum's still boiling over from the Z5 mobile results.
I'd be the advocate of waiting for some likely fine-tuning in the next months, and it looks like a solid improvement in power consumption for an equal perf over Z4...but at the same time what's the point of bumping 8 Z4 cores to 12 Z5 cores if you're going to have the same perf, lel.
AMD clearly fumbled something. I might've considered that with a sufficient amount of power we'd see something stronger, but no. That's the laptop they sent to reviewers, it's likely that that's what the product's meant to be.
The option "AMD splurged on the core count to get some leeway with reviews" also isn't viable, at least not in this dimension. AMD never splurges.
That's the crippled mobile part is it not?According to David Huang it's not even double digits performance increase in Specint.
I'd argue 2 new cores is much more of a difference than just 1.Node wise, Intel has the bigger jump, but core wise, Zen 5 seems like a much more radical departure than LNC is over GLC. And tbh, some parts of LNC don't seem to be buffed as much as a usual Intel tock is, like the ROB capacity or the uop cache capacity. I don't think the same can be said for Zen 5.
It's not like Intel tried for a massive core arch refresh and got mid results, they literally didn't try at all.
I would consider Zen 5 a bigger "meh" than LNC, but as a whole, the entire "which core is more disappointing vs expectations" shtick isn't really important in the end lol.
To save power, same strat as every big.LITTLE part ever.Splitting core to 2 different CCXs really hurt performance. Still don't understand why they did it, it seems that doing one 8-core hybrid CCX solution would have been better alternative.
Bulldozer also had horrible power and area eff on top of being slow.Regressing in the GCC subtest after all the accolades about that zero-bubble, 2-branches BPU with 16k L1 BTB... Jesus, this is Bulldozer vibes.
You can look at the specs that are out for LNC dude, they did not cram 2 cores worth of improvements into LNC. I genuinely think they barely did one. It's not like Intel increased the ROB entry capacity by 100%, doubled decoder width, and doubled the width of the renamer lol.I'd argue 2 new cores is much more of a difference than just 1.
So what's with the much higher power consumption with ST workloads on AMD/Intel APUs? Is the core power that much higher or uncore?
Any chance of being fixed in future generations?
View attachment 104096
Where is this chart? I'm not finding it within https://www.notebookcheck.net/Asus-...en-5-inside-a-1-3-cm-thick-case.868219.0.htmlSo what's with the much higher power consumption with ST workloads on AMD/Intel APUs? Is the core power that much higher or uncore?
Any chance of being fixed in future generations?
View attachment 104096
Premium you mean, collectors itemThey could have just sold the mislabeled 9700x with $25 discount.
All those numbers are inaccurate, better to look here for more precision :
AMD Ryzen AI 9 HX 370 „Strix Point“ im Zenbook S16 im Test
AMD „Strix Point“, die neue APU mit Zen-5(c)-Kernen, RDNA 3.5 und starker NPU, ist da. Im Test ist der Ryzen AI 9 HX 370 sehr effizient.www.computerbase.de
Where is this chart? I'm not finding it within https://www.notebookcheck.net/Asus-...en-5-inside-a-1-3-cm-thick-case.868219.0.html
How are they testing 80W for the Ryzen AI? Didn't think the sku supported this power profile?View attachment 104097
Interesting image I saw elsewhere
Figures I've saw being throwed some time ago was around 170 - 180mm²Does anyone have rumors or estimate of the die size of Kraken Point?
It comply with the Microsoft marketing point, it will be featured into newer designs and with more design wins, it will have a branding that indicates it's a new generation part and will also slot into a cheaper price bracket than HWK.From a customer perspective, why would anyone buy Kraken Point over Hawk Point?
Numbers are different (lower), but trend is quite similar. M3 parts consume 3-4x less power for ST compared to Strix/Hawk/meteor. At least Strix looks better than Hawk point.
They have a new article: https://www.notebookcheck.com/AMD-Z...und-Qualcomm-Snapdragon-X-Elite.866997.0.html
I said there's an argument to be made, not a definitive answer. However, in the grand scheme of things, the difference between raptor and arrow is more than zen 4 and zen 5.
True, I didn't catch the initial spike.M3 consume much less because it throttle due to temperature, otherwise it use 10W in ST for 10 seconds at the begenning of the run.
The 370 use 18-20W in ST and the 155H 25-28W.
True, I didn't catch the initial spike.
So I guess 2x worse ST power consumption for Strix. I guess that's not as bad (especially if M3 Pro etc are higher, plus N3B). Also promising that Strix has lower consumption compared to hawk point with better performance too. Maybe next gen will be even more competitive!
The peak ST power consumption will be 10 watts for all M3 series chips, they all clock at 4.05GHz.ST power consumption for Strix. I guess that's not as bad (especially if M3 Pro etc are higher, plus N3B
Perhaps. I was basing my comment off NBCs measurements, which show higher ST power usage (but same score) for M3 pro and M3 max compared to M3. I rationalized this as worse power gating (of course I might be wrong, but bigger die for sure).The peak ST power consumption will be 10 watts for all M3 series chips, they all clock at 4.05GHz.
Why Is the 9950x the only zen 5 part that looks impressive probably because all the others reduced tdp
Getting this result on an air cooler is impressive. 13/14th gen needs an aio
The STX in Asus ProArt can go upto 80W.How are they testing 80W for the Ryzen AI? Didn't think the sku supported this power profile?