I might have the numbers wrong, but the 2 i'm comparing and complaining about, i really dont see any size reduction. just a different ( incompatible ) shape. and lower wattage.
Sure, if you saved a bunch of space, or it was magically waterproof or something, id say its 'ok' but i just dont see it here. I think when i buy transformer dongles from here out out ill be sure check first to avoid it.
And yes i know its low on the list of things to be pissed about, but getting really tired of stuff like this "just because we can make you buy new stuff" .
NVIDIA's days are numbered. Generic TPUs are coming for them, and now this -> https://siliconangle.com/2024/07/18/openai-reportedly-holding-talks-broadcom-others-develop-new-ai-server-chip/
( broadcom and openAI are in talks.. )
Sort of funny. Intel 13 and 14 gen core series CPUs crash on load. So does gen 12. ( i can attest to that personally... i got mine stable finally, after disabling a lot of things, including all the 'power saving cores' so i lost 1/2 of what its supposed to be able to do )
Intel sux they need to get out of the way.
https://www.techspot.com/news/103876-intel-laptop-cpus-crashing-company-remains-silent.html
Are you running an i9 or something less? Apparently these problems affect the whole lineup, but they are worst with Thermal Velocity Boost (i9-only feature)
I've been thinking about upgrading to Arrow Lake when it's available, but with these issues unresolved I'm not so sure. Zen 2 it is, for now.
An i7. 12gen. Shoved in a NUC like box and at the time was a bit faster than most without spending silly money. ( and why i chose it ) Its a couple of generations old now as I put it away for almost a year as i was too frustrated and came close to tossing it. I had read about other people with similar issues with that gen, mainly related to core graphic drivers and such. None of the stuff they suggested worked ( for many people it seems, not just me ) so away it went. When i brought it back out during some house cleaning, "lets try this again, but at BIOS level, before i finally toss it" mostly i disabled the 'efficiency cores' which of course would drop power use, so im 99% its related to the current issue even tho at the time i didn't know about that. The other things i messed with, i doubt they really made a difference. ( mostly pre-fetch and extension stuff, i was just trying random things, and most i turned back on anyway )
Disabling the cores sort defeated the use-case i had bought for it, now its 'yet another mid-range i7' but at least its not collecting dust and i can use it.
Thu Jul 25 2024 13:49:09 EDT from LoanShark
Are you running an i9 or something less? Apparently these problems affect the whole lineup, but they are worst with Thermal Velocity Boost (i9-only feature)
Hmmmm. I have had faulty hardware before. In the 90s, my brother bought an AMD K5 (or something like that) which had stability issues in Linux at nominal voltage and frequency settings. (As soon as you ran the `df` command, it would crash. Every time.) Thesse problems went away when you jumpered it down to a lower frequency. But I think they eventually came back, even at the lower freq - the hardware was degrading before our eyes.
Maybe some more recent examples, too, but that's the one that stands out in my mind.
Supposedly, the 12th gen doesn't suffer from the SAME issues as 13th/14th. And apparently, they've finally root-caused the latter: https://www.notebookcheck.net/Intel-breaks-silence-on-13th-and-14th-gen-Raptor-Lake-desktop-CPU-instability-issues.866224.0.html
Ya i tried stuff like reducing clock speeds and such early on. No effect. Also thought network for a while too, but no, that wasn't it.
I forget the exact errors i got on console when it didn't freeze solid but all running processes did, and it always seems to be related to one or 2 of the cores, so i thought 'why not give that a try and turn those off" It was the 'power saving' ones, ( whatever its called ) so i was just assuming its power related after reading the stuff about 13 and 14 gen. I also don't think it was bad hardware, as i was not the only person, and they had drastically different hardware, but same CPU...
Did not read the entire article as im busy, but apparently the 'fix' intel is putting out only apples if you have not had issues. If you have, your CPU is physically damaged. ( it was on the verge if anyone wants to look )
"If you have experienced crashing your CPU is irrecoverably damaged and should be replaced"
And they are still blaming voltages, and their microcode doing it.
Also who is paying for these replacements? And what about laptops and NUCs, where you cant normally replace it?
lol
Fri Jul 26 2024 16:35:19 EDT from LoanShark
Why do you suppose they are denying that laptops are affected? ;-)
booooo my data ( not system ) drive died in my main PVE server. The one that houses the NAS VM data...
4TB SSD . Poof. Even caused a hardware reboot.
ya ya i know.. backups, ceph, bla bla but other than occasional backups, anything else is overkill
Weird stuff afoot. Restore completed 13 hours. Not that the VMs change much, other than the NAS beast so i dont *need* to, but went ahead and doing some updated backups/etc anyway. Backups from host server 1, most were running at 250m/s or so. ( not yet backed up a new copy of the beast. still getting all the media updated first. backup was perhaps 4 months old so missing things )
Host server 2 ( mostly for testing/etc, or when i want dedicated horsepower ) was going to run a backup of a Debian desktop i use for incoming RDP. Same story, dont 'need' it, but figured might was well since the backup server is plugged in. Doh, i numbered it wrong. Did a full clone so i could reset the ID. 300m/s ( ok, not great but not terrible, but to be fair its all on a 2TB m.2 in that box, SSD is just for 'stuff'. ). Back it up. Started out at 230m/s, and quickly dropped to 80m/s. Sometimes even dropping down to 15m/s. Then back up, then down.. then sideways.. Ran much better and more consistent last time i did it. ( and same here no vms running to reduce contention )
wtf.
( and i guess i should be happy it works. and that i have the stuff to bitch about in the first place. i have more than many. Perhaps ill just shut up. )
Well, looks i get a replacement drive from the manufacturer 3 year warranty. It was only a couple of days: "we are sorry it took so long to get back with you we have been swamped, here are the steps to get an RMA and send it back"
Point in their favor.
Are you running an i9 or something less? Apparently these problems
affect the whole lineup, but they are worst with Thermal Velocity Boost
(i9-only feature)
Gee, now I'm glad I disabled boost on my i9 :)
(Even though I did it to reduce excess power consumption ... "not crashing" is a nice feature.)
Context not important, but got a flyer for a 'new and improved portable model' Among other "advancements" ( like a microsd port *yawn* ) it includes: "179% faster processing: powered by the IntelⓇ Processor N100"
If a POS N100 is 180% faster i would hate to see what it used to be like ...
So the drive that died and i sent back, they just shipped the replacement. No questions asked ( not that i did anything to cause it.. but you know how some companies are ). Will have to give them credit for standing behind their products.
That is a bit of cash...
"Advanced Micro Devices will acquire server manufacturer ZT Systems for $4.9 billion as the chipmaker looks to expand its artificial intelligence capabilities."