- Mar 3, 2017
- 1,779
- 6,798
- 136
cool story bro have fun with zen5 halo till 2028If intel PTL won over Halo anywhere, Intel would plaster that everywhere. They didn't.
The graph you quote below is Halo 395, which is literally a 9950X (literally) running on 15W with the fattest iGPU running Cyberpunk... 15W total and still higher than stxpoint.
Halo 388 (40CU + 8 cores instead of 16) should have better performance but also handhelds can do 25W+ too
Price wise, Halo 388 (32GB) has higher performance than PTL and should have similar price
Handhelds will be using a different SKU called G3, and Intel hasn't shown performance numbers for that.If intel PTL won over Halo anywhere, Intel would plaster that everywhere. They didn't.
As opposed to what?cool story bro have fun with zen5 halo till 2028
nothing. There is no comp in x86 land because Intel as always is too slow to respond.As opposed to what?
Uh, no.At least by the end of 2027 Intel will finally release Razor Lake-AX
Medusa AT4/AT3 aren't releasing in early 2027?That doesn't mean I'm happy with AMD not making a Zen6 mobile Halo anytime before 2028.
However, for 15-20W handhelds I still expect Panther Lake's G3 series to come out on top of Halo
leaked roadmaps had only medusa point-1 in early 2027 & medusa-point 3 (bumblebee) in late 2027Medusa AT4/AT3 aren't releasing in early 2027?
Why shouldn't we celebrate another player?Wow, in the super low power segment where Intel were already dominant, and expensive hand held gaming, which is a super niche, Intel is now dominant. That gets them... no market share. Well done Intel!



Wow, in the super low power segment where Intel were already dominant, and expensive hand held gaming, which is a super niche, Intel is now dominant. That gets them... no market share. Well done Intel!
Another player in what? There are already handhelds running in the 25W power envelope, and the 388H is a low volume part.Why shouldn't we celebrate another player?
At least "expensive" part should get better. Intel is the underdog in this segment ...
Yes, Phawx seems to be getting worked up over nothing.Imagine getting this much worked up because Intel managed to score a clear win.
my problem is that there is only one player. TSMC for GPUsAnother player in what?
View attachment 137349
View attachment 137348
View attachment 137347
These are all results without upscaling. XeSS 3 vs. FSR3 at ISO IQ would probably increase the advantage to Intel's side.
Imagine getting this much worked up because Intel managed to score a clear win.
Small difference on a more expensive platform from AMD is a win for Intel.- Small diff
Feel free to provide other examples.- Cherrypicks (barely any diff on doom TDA)
Not really. The second CCD on Strix Halo gets into sleep / deep-c state below 20W, so power efficiency between the 395 and the 388 at that power level should be similar.- Halo 388 should run faster leaner at low W
Halo's IOD is almost twice as big as PTL-H's 3 chiplets combined, and then it needs the CCD plus twice the memory chips.- PTL price won't be far from Halo 388
CPU-wise, Panther Lake will beat Halo's single CCD 388 in MT (which is also beat by the 12-core Strix Point) and should be neck-and-neck in ST.- Halo has real full flagship CPU of high-end desktop PC... 9700X / 9950X
That's the only advantage of Halo, assuming it's implemented in a system that can actually cool down 65W.- HALO DOCKED MODE unlocks 65W, cannot compare
I doubt the 388H is coming to handhelds. Intel is launching the G3 and G3 Extreme for that market.These benchmarks only shows that PTL 388H is a better chip for handhelds than stx Halo. And that was pretty obvious before since stxH is a 55w part.
AMD handheld offerings are dead.even if there's no more Z-series from AMD anytime soon unless Z3 is based on Halo.
AMD handheld offerings are dead.
It's not "rapidly rising" at all.It'll be both parts ironic and depressing to see AMD falling behind on a rapidly rising market whose hardware they pretty much gave birth to.
If your average user already own a 9800X3D it makes very little sense to buy a 9850X3DSomething wrong / unoptimized with the setup if a 9850X3D loses to a 9800X3D
Looking at this from a purely overclocking / hwbot lens:
9850X3D is a superbinned 9800X3D, how much better the silicon quality on the CCD is, depend on what sample you are comparing it against.
The very best 9800X3D can probably match it / get close, but there are very few of this caliber out there in the wild.
Average sample vs average sample i would say + ~200-300mhz clockspeed once overclocked
(difference is larger when running stock 5.2ghz vs 5.6ghz like reviewers do)
So the 9850X3D is just a way to get very strong silicon, without binning hundreds of 9800X3D's
IO-die quality (IMC/FCLK) is still all random, so no guarantee there.
If your average user already own a 9800X3D it makes very little sense to buy a 9850X3D
These new cpus are for overclockers or ppl that are building new systems.
and... creating 3 slides a year and 2 pages on amd.com where you just copy paste spec of some anyway for notebook delivered SKU just to signal that you are "committed to handheld gaming" is not worth it?It's not "rapidly rising" at all.
Maybe 2m units total a year.
It's not "rapidly rising" at all.
Maybe 2m units total a year.
it will be huge. in 3 to 4 years time a major chunk compared to first party xbox sales30% a year is rapidly rising.
It was 2.3M in 2025, projected ~3M in 2026 and 3.9M in 2027. At an average ~$800 per unit, it's a sizeable sub-market within the gaming PC ecosystem (same games, same peripherals, etc.).
