|
Yudo posted:Leaky chips that can clock = consumer desktop. Scalability and reusablity is the raison d'être of AMD's designs, with good reason. Or, at least, that is what I thought. The basic idea is right, there's design and physical re-use up and down the stack. But a host of technical and logistical issues make it more complicated than "this CCD can't cut it as a server or laptop, stick it in a desktop". At the 10,000 foot level there are tradeoffs that can be made for more or fewer desktops vs fewer or more servers, but desktop can't subsist on server's rejects.
|
#
?
Jun 11, 2023 02:56
|
|
|
|
| # ? May 30, 2024 17:00 |
|
|
It's more true from an R&D sense than a manufacturing one I think
|
|
#
?
Jun 11, 2023 02:59
|
|
|
The return of non-pro threadrippers? https://twitter.com/momomo_us/status/1667173707217874946 edit: the thumbnail crops out the most relevant part: 
Dr. Video Games 0031 fucked around with this message at 04:02 on Jun 11, 2023 |
|
#
?
Jun 11, 2023 04:00
|
|
|
Dr. Video Games 0031 posted:The return of non-pro threadrippers? They're on SP6, so they'll be based on Zen4 Siena, which is decently cut down from the full-size Threadripper Pro on SP5. It's a similar split to W2400 vs W3400 Xeon-W, you get lower core counts, half the memory channels and less PCIe. Siena also caps out at 225W, which is a bit of a limitation for 64core cpus, but I dunno if that is a socket hard cap that will actually limit Threadripper in any meaningful way. It'll still be cool though. Apparently they're dropping the OC support they added with Zen3 TR Pro and that'll be TR non-Pro exclusive again.
|
|
#
?
Jun 11, 2023 10:33
|
|
|
Guess I might be getting myself a 7800X3D after all because ASUS couldn't fix my X570-I and the retailer is refunding the motherboard. Cooool. The motherboard that was £270 when I got it in 2020 and is now like £350 assuming you can even find it new anymore. Guess I'll find my partner a respectable ATX board and upgrade the 2600K finally.
|
|
#
?
Jun 12, 2023 10:55
|
|
|
Probably stupid question, but is there a reason to use the chipset drivers that come from the mobo manufacturer over the ones provided directly by AMD themselves? e: for desktop motherboards
|
|
#
?
Jun 12, 2023 23:44
|
|
|
Kibner posted:Probably stupid question, but is there a reason to use the chipset drivers that come from the mobo manufacturer over the ones provided directly by AMD themselves?
|
|
#
?
Jun 13, 2023 00:03
|
|
|
I have almost managed two weeks of uptime on my previously ridiculously unstable 7700X + ASUS B650 build. I don't care that this post is going to jinx that. I'm just amazed at this because I haven't changed a loving thing. No BIOS updates, no physical environmental changes, no driver updates, same set of Windows updates still outstanding... This machine is more haunted than myhouse.pk3. e: Crashed six and a half hours after this post. loving pile of poo poo. Kazinsal fucked around with this message at 18:17 on Jun 13, 2023 |
|
#
?
Jun 13, 2023 10:35
|
|
|
Did I imagine it or was there talk of a 7600X3D?
|
|
#
?
Jun 13, 2023 20:54
|
|
|
Yeah it’s being spoken of as a foregone conclusion as one last ride for AM4 since they probably have some binned 5800X3Ds lying around.
|
|
#
?
Jun 13, 2023 21:12
|
|
|
ijyt posted:Did I imagine it or was there talk of a 7600X3D? If this isn’t a typo, the existence of the 5600x3d would seem to indicate it might happen eventually but I don’t think anyone is making noise about it?
|
|
#
?
Jun 13, 2023 21:35
|
|
|
AMD released the 3D v-cache and 128 core epycs today. First party comparison between the 3D v-cache chips and sapphire rapids max (i.e. 64GB of in-package HBM): 32 core vs 32 core:  Max cores each company offers:  There's also their AI accelerator https://www.anandtech.com/show/18915/amd-expands-mi300-family-with-mi300x-gpu-only-192gb-memory. ijyt posted:Did I imagine it or was there talk of a 7600X3D? https://twitter.com/g01d3nm4ng0/status/1667502118456918016
|
|
#
?
Jun 13, 2023 21:39
|
|
|
5600X3D seems like a very weird product when the 5800X3D is $280. Like if you want to buy your end cap AM4 cpu, you prolly want the last 2 cores to last you as long as possible. And if you want the best bang/buck upgrade for an old build, a 5600 is $119. Is there a significant market in between? Maybe China focused? I dunno, seems like a strange late launch.
|
|
#
?
Jun 13, 2023 21:58
|
|
|
Cygni posted:5600X3D seems like a very weird product when the 5800X3D is $280. Perhaps they have stacks of ccxs that don't make the 5800x3D cut. Chasing less affluent markets is an interesting idea. AM4 is cheap, stable, and parts are still very much available.
|
|
#
?
Jun 13, 2023 22:14
|
|
|
I just noticed in that last screenshot - It’s funny that they knocked 100MHz off the 5800x3d’s clock. I was wondering if the 5600x3d might have a slight advantage due to thermals. Or maybe the CCDs for the 5600x3d are terrible bins that need all of the max allowed voltage for a zen 3 vcache for 6 cores.
|
|
#
?
Jun 13, 2023 22:24
|
|
|
AM4 continues to live at the budget end. My guess is interest in 5800x3d is slowing so they're not eating their own lunch if they target the next value segment down.
|
|
#
?
Jun 13, 2023 22:33
|
|
|
hobbesmaster posted:If this isn’t a typo, the existence of the 5600x3d would seem to indicate it might happen eventually but I don’t think anyone is making noise about it? Ah must have been the 5600X3D and my memory flipped a bit lol, nvm!
|
|
#
?
Jun 13, 2023 22:55
|
|
|
ConanTheLibrarian posted:AMD released the 3D v-cache and 128 core epycs today. First party comparison between the 3D v-cache chips and sapphire rapids max (i.e. 64GB of in-package HBM): These benchmarks are against regular Sapphire Rapids; the Xeon Max CPU are 94xx, not 84xx. So no HBM.
|
|
#
?
Jun 14, 2023 09:58
|
|
|
Well poo poo.
|
|
#
?
Jun 14, 2023 20:18
|
|
|
Cygni posted:5600X3D seems like a very weird product when the 5800X3D is $280.
|
|
#
?
Jun 14, 2023 21:25
|
|
|
stacking happens after binning, and there are no defects that are going to happen during/after stacking that are going to leave you with a functional chip in any respect, if it fails during stacking it's dead. just a reminder but all epyc v-cache chips are 8-CCD with 768MB of cache on them. so for example a 7473x is 16C, and that means they're stacking v-cache on 2-core CCDs. Since AMD makes 16, 24, 32, and 64c models, there are epyc binning streams for 2c, 4c, and 8c CCDs, but not 6c ones like 5600x would use, since there is no 48C epyc v-cache SKU. if there are too many 5800X3D chips made, or too many epyc 8C dies stacked, they could be binning down 8c chips into 6c ones for salvage - just like they turned 7900X into 7600X and 7950X into 7800X. CPU sales (especially consumer sales) are miserable overall and AMD is having to shuffle inventory around to the SKUs that are actually selling. Or they may have excess space on the assembly line and hey, we're selling every 5800X3D we can without dropping prices too far, add a new 5600X3D sku instead. but it's not a case of core harvesting or clock harvesting or epyc defect harvesting in general. it's rare for a chip to fail that late, we are talking sub-1% in general, but when they do fail the whole chip is dead at that point. Same for the dual-CCD 7600Xs/etc - there is no meaningful amount of 7900X that fails during packaging and results in a viable IO die and CCD, such that two random youtubers get one within a week. at the current price it's pointless, but if it drops to $200 there's a market for that. It'd be an esports champ. the other possibility is since 5800X3D is getting so far below MSRP that AMD is going to slot a new SKU and stop dropping 5800X3D so much, so it may be a case of "your choices are now 5600X3D at $250 or 5800X3D at $329" rather than $250-vs-$280.
|
|
#
?
Jun 14, 2023 22:07
|
|
|
What piques my interest in the 5600X3D is its TDP of 65W. You could have it in much smaller systems or perhaps it'll even be packaged with its own cooler once again. It obviously won't be as fast as the 5800X3D but it could end up being a massive bang-for-buck king. Zedsdeadbaby fucked around with this message at 07:52 on Jun 15, 2023 |
|
#
?
Jun 15, 2023 07:50
|
|
|
Zedsdeadbaby posted:What piques my interest in the 5600X3D is its TDP of 65W. 7700 is 65W and as far as I recall is faster in most games than a 5800X3D. I built my htpc with a 7700 for that reason
|
|
#
?
Jun 15, 2023 09:47
|
|
|
Zedsdeadbaby posted:What piques my interest in the 5600X3D is its TDP of 65W. It could be a neat budget option depending on the pricing but you can already cap the 5800X3D at 84 PPT for the same power efficiency.
|
|
#
?
Jun 15, 2023 11:31
|
|
|
HalloKitty posted:7700 is 65W and as far as I recall is faster in most games than a 5800X3D. I’d like to extend the life of my stupid low power AM4 build so it’s a really nice option for that!
|
|
#
?
Jun 15, 2023 12:03
|
|
|
KYOON GRIFFEY JR posted:I’d like to extend the life of my stupid low power AM4 build so it’s a really nice option for that! In an existing am4 it's no doubt going to be attractive
|
|
#
?
Jun 15, 2023 12:37
|
|
|
Has anyone gotten any of those Noctua off set plates for the 7000 series? Any feedback?
|
|
#
?
Jun 18, 2023 04:56
|
|
|
Should I be concerned with a 5800x3d hitting ~86c as the max temp during long gaming sessions? (total war warhammer 3) - average was ~79c according to hwinfo. Trying to decide if I should just be done with upgrades or if I should swap out the tower cooler for a newer one.
|
|
#
?
Jun 18, 2023 17:34
|
|
|
5800X3D does weird things. easiest thing is to use the curve optimizer to lower it by (-)20 the temps are going to be weird and spike in a lot of cases on account of it being experimental and perhaps a big variance in thermals across chips if you're going to replace a cooler, it might not be the worst idea to go with watercooling, because my d15 still does funky things. i've moved around some fan curves, increased the hysteresis (ramp time) and figured out what level of noise is ok for higher loads, and it runs fine, but it still gets a bit hot from time to time. ~80 for cinebench multicore which serves as a good stress test
|
|
#
?
Jun 18, 2023 17:56
|
|
|
You also could try just updating your motherboard BIOS; mine was ostensibly already on a version that supported the X3D chips but it took an update to the absolute latest version to solve some of the "straight to throttling range under even light workloads" issues I was seeing, Nowadays it rarely goes above 75C on a -20 UV
|
|
#
?
Jun 18, 2023 18:08
|
|
|
also always update your chipset driver!
|
|
#
?
Jun 18, 2023 18:22
|
|
|
fargom posted:Should I be concerned with a 5800x3d hitting ~86c as the max temp during long gaming sessions? (total war warhammer 3) - average was ~79c according to hwinfo. Trying to decide if I should just be done with upgrades or if I should swap out the tower cooler for a newer one. Gaming? Those are pretty high for games, it would be understandable if it was cinebench or something but most games aren't going to peg all cores like that.
|
|
#
?
Jun 18, 2023 21:42
|
|
|
fargom posted:Should I be concerned with a 5800x3d hitting ~86c as the max temp during long gaming sessions? (total war warhammer 3) - average was ~79c according to hwinfo. Trying to decide if I should just be done with upgrades or if I should swap out the tower cooler for a newer one. Any half-decent tower cooler should do an adequate job on a 5800x3d -- the CPU generates fewer watts of heat than a regular 5800X. But they get hotter than the regular because the 3d silicon is a blanket. I would look at 1) your case ambient and whether you need to address overall airflow and 2) if you may have mounted the cooler poorly or just need to tighten it down.
|
|
#
?
Jun 18, 2023 22:31
|
|
|
My 5800X3D runs about 62C in games, give or take, with a cheap 120mm air cooler. Now if I run Prime95 it goes straight to 90
|
|
#
?
Jun 19, 2023 02:15
|
|
|
Frequency is as important as temperature because zen 3 (all am4?) start reducing clocks slightly for every degree above something pretty low, like 60C.
|
|
#
?
Jun 19, 2023 03:17
|
|
|
hobbesmaster posted:Frequency is as important as temperature because zen 3 (all am4?) start reducing clocks slightly for every degree above something pretty low, like 60C. This is true but: 1) the X3D has a limited, lower maximum boost than other zen 3 chips 2) even in non-X3D chips it's pretty dang trivial. Like, water cooling at 50c will get you ~50 mhz more than a 120mm heatpipe like a U12S or whatnot at 65-70c. (cite: GN water vs air video from a year ago) Mid-80s is abnormally hot though, and might be more impact. edit: or if you mean, OP's chip might be running 300 mhz under cap, yeah. that's a good question.
|
|
#
?
Jun 19, 2023 03:39
|
|
|
Klyith posted:edit: or if you mean, OP's chip might be running 300 mhz under cap, yeah. that's a good question. Yeah.
|
|
#
?
Jun 19, 2023 04:37
|
|
|
new mega mobo benchmark just dropped https://www.youtube.com/watch?v=ZtHOOyWYiic
|
|
#
?
Jun 24, 2023 12:19
|
|
|
kliras posted:new mega mobo benchmark just dropped
|
|
#
?
Jun 24, 2023 12:47
|
|
|
|
| # ? May 30, 2024 17:00 |
|
|
Cygni posted:5600X3D seems like a very weird product when the 5800X3D is $280. i have zero desire to upgrade from my existing 5600
|
|
#
?
Jun 24, 2023 13:01
|
|
