Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us $3,400 per month for bandwidth bills alone, and since we don't believe in shoving popup ads to our registered users, we try to make the money back through forum registrations.
«432 »
  • Post
  • Reply
Cygni
Nov 12, 2005

raring to post



priznat posted:

Also I hate m.2 on anything other than a laptop.

now thats the kinda badpinion i crave from a message board

Adbot
ADBOT LOVES YOU

priznat
Jul 7, 2009

Let's get drunk and kiss each other all night.

It is hard won from sticking dozens of optane 16GB modules on aic adapters and those loving screws... Tiny stupid stubby screws!! Gah I hate them.

Motherboard makers who use the plastic retention clips are the best u supermicro

Statutory Ape
Sep 12, 2017



priznat posted:

The form factor is the worst! Give me aic or u.2 anyday.

i will grant you that its perhaps not an ideal shape against a desktop mobo


otoh, running a cable.

priznat
Jul 7, 2009

Let's get drunk and kiss each other all night.

Just do an aic form factor, pcie slot, done. Lots of room for extra caps for power event handling and oodles of nand. Better thermals too.

Taking something designed with size constraints and using it someplace that doesn’t have those constraints is silly imo.

Manufacturers love m.2 though because they can crank them out for both mobile and desktop users so that’s what we’ll get

Volguus
Mar 3, 2009


priznat posted:

It is hard won from sticking dozens of optane 16GB modules on aic adapters and those loving screws... Tiny stupid stubby screws!! Gah I hate them.

Motherboard makers who use the plastic retention clips are the best u supermicro

If one would have to do this every day, especially on an already built PC, yes I agree, those little screws can be annoying. Though, I presume that if one has to do this every day, then said one would equip himself/herself with some magnetic screwdrivers or some other tools to make life easier.

But since normal people don't have to change their M2 drives very often, it's not a problem for more than 10 minutes.

CaptainSarcastic
Jul 6, 2013

HAIL SATAN


I feel pretty neutral about m2 drives on desktops but that is likely because the only time I've ever installed them is on my new build where I put them in before I installed the GPU. It would be a pain in the rear end to do now, but as part of the initial build process it was fine.

Methylethylaldehyde
Oct 23, 2004

BAKA BAKA


We really just need more pcie add in boards that let us put 4 or 8 m2 drives in one 4x or 8x slot.

Daddy needs him some nvme software raid. Like 10 tb of software raid.

Oxyclean
Sep 23, 2007




CaptainSarcastic posted:

I feel pretty neutral about m2 drives on desktops but that is likely because the only time I've ever installed them is on my new build where I put them in before I installed the GPU. It would be a pain in the rear end to do now, but as part of the initial build process it was fine.

Same. In terms of a fresh build the m2 was way less hassle. Stick in, screw in, done.

Plus it's not like regular drives have been hassle-free for me in any circumstance. I wanted to stick a new SSD in an old build and my GPU was blocking all the SATA ports. Idiot me totally forgot where to press to release the PCIE slot for way too long and spent a bunch of time worrying I was about to break something. Then I forgot to hook up power to the drive because I was thinking "oh that's like, what you had to do for older drives."

Also cable management. I'm so happy my current case has really nice cable management by basically banishing all the PSU cords into the basement. Downside was threading the main mobo power felt like I was on the verge of making the cable bend in a way it shouldn't.

in other words less cables is more good.

movax
Aug 30, 2008



Methylethylaldehyde posted:

We really just need more pcie add in boards that let us put 4 or 8 m2 drives in one 4x or 8x slot.

Daddy needs him some nvme software raid. Like 10 tb of software raid.

Agreed — all it takes a PCIe switch!

Also — obviously the forums are on fire to some degree, if you want to post about it, check this out here: https://forums.somethingawful.com/s...hreadid=3929022

Crunchy Black
Oct 24, 2017

CASTOR: Uh, it was all fine and you don't remember?
VINDMAN: No, it was bad and I do remember.




I'm just waiting for someone to have a truly beautiful take in the AMD thread like, "ruler form factor is good." Cause that's the kind of day we're having on these fair forums.

Paul MaudDib
May 2, 2006

"Tell me of your home world, Usul"


movax posted:

Agreed — all it takes a PCIe switch!

Also — obviously the forums are on fire to some degree, if you want to post about it, check this out here: https://forums.somethingawful.com/s...hreadid=3929022



https://highpoint-tech.com/USA_new/...-1-overview.htm

movax
Aug 30, 2008




Now we just need the PEX8896 (or w/e they call it) 96 lane PCIe 4.0 switch to bandwidth bridge to PCIe 3.0 NVMe drives because that's still going to be so loving fast.

Methylethylaldehyde
Oct 23, 2004

BAKA BAKA



I really want it, but I also wanna pay rent this month

Volguus
Mar 3, 2009



drat, RAID 0 on 4 M2s that should be quite the (theoretical) speed. And it's "only" CDN$ 562.91 on amazon. I would have expected it to be quite much more, but at this price I can dream about it.

priznat
Jul 7, 2009

Let's get drunk and kiss each other all night.

movax posted:

Now we just need the PEX8896 (or w/e they call it) 96 lane PCIe 4.0 switch to bandwidth bridge to PCIe 3.0 NVMe drives because that's still going to be so loving fast.

PFX PM40100 you mean

Also make sure to manually set the slot for “preferred IO” given that this is the AMD thread

Wrar
Sep 9, 2002



Soiled Meat

Does anyone have any idea when x570 boards will return to something resembling availability/price sanity? I have the upgrade itch.

DOOMocrat
Oct 2, 2003



The current rule of PC building is:

Case, board, CPU, ram, power supply, storage.

Four of these six things will be in stock this week. The other two aren't. I would expect AMD to be pushing a stock refresh around the time of the XT launch but I'd set you some Newegg alerts if what you have isn't up and on sale from them.

Pablo Bluth
Sep 7, 2007

I've made a huge mistake.


Stock levels are fine this side of the Atlantic.

SwissArmyDruid
Feb 14, 2014



Crossposted from the laptop thread:

Hadlock posted:

re: intel vs amd

https://www.pcgamer.com/intel-skylake-why-apple-left/

tl;dr intel has been in a bad place for a while, they have been through what, 3 CEOs since 2017?

Skylake is uh, 2 generations prior to the thread namesake Haswell, which debuted in 2012, so this is old news, but sort of confirms a slowly gelling story of things going to poo poo over the last 8 years

also RIP laptop megathread, I will miss all of you, from arguing that laptops were not an acceptable gaming platform, to uh, wether or not eGPU was the superior laptop gaming platform. I will cherish these banal arguments forever, long live Thinkpad Refurbishments

quote:

That makes complete sense from a business and an architectural point of view, but while it's something Piednoël says was always under consideration by Apple, he believes if the company hadn't found so many issues within the Skylake architecture it would still be onboard the Intel chip train. It was the straw that broke the Apple's back, so to speak.

"The quality assurance of Skylake was more than a problem," says Piednoël during a casual Xplane chat and stream session. "It was abnormally bad. We were getting way too much citing for little things inside Skylake. Basically our buddies at Apple became the number one filer of problems in the architecture. And that went really, really bad.

"When your customer starts finding almost as much bugs as you found yourself, you're not leading into the right place."

[...]

"For me this is the inflection point," says Piednoël. "This is where the Apple guys who were always contemplating to switch, they went and looked at it and said: 'Well, we've probably got to do it.' Basically the bad quality assurance of Skylake is responsible for them to actually go away from the platform."

Intel ruining things for AMD too.

SwissArmyDruid fucked around with this message at 19:11 on Jun 25, 2020

Fabulousity
Dec 29, 2008

= (Displacement through a hetero medium) / Time


Nap Ghost

I take this to mean that Intel is also going bananas on the new corporate fad of asphyxiating quality assurance and control resources in the name of improving project timelines and decreasing development costs in the short term with zero thought towards the long term and non-obvious consequences of doing that?

I guess they aren't wrong that there are no (documented) bugs if you don't test

Methylethylaldehyde
Oct 23, 2004

BAKA BAKA


Fabulousity posted:

I take this to mean that Intel is also going bananas on the new corporate fad of asphyxiating quality assurance and control resources in the name of improving project timelines and decreasing development costs in the short term with zero thought towards the long term and non-obvious consequences of doing that?

I guess they aren't wrong that there are no (documented) bugs if you don't test

"All these QA assholes do is bring me problems. I don't wanna hear about problems, I wanna hear about solutions!" *Fires half of them in a fit of pique*.

**3 years later spongebob meme**

"What do you mean all our products have horrible, terrible no good very bad bugs in them? What is QA even doing?" *Fires the rest of them, outsources it*

Riven
Apr 22, 2002


Fabulousity posted:

I take this to mean that Intel is also going bananas on the new corporate fad of asphyxiating quality assurance and control resources in the name of improving project timelines and decreasing development costs in the short term with zero thought towards the long term and non-obvious consequences of doing that?

I guess they aren't wrong that there are no (documented) bugs if you don't test

All this testing just keeps finding more cases!

WhyteRyce
Dec 30, 2001



Silicon validation is highly disrespected. People outside the industry think you just run benchmarks and game demos over and over again as a low level entry job. companies in the industry hate you because you're the guy holding up release and demand a huge amount of resources that can't be monetized or productized so it's just pure money down the drain to them. Program managers hate it because it either generates noise/false flags or finds weird corner cases they don't even know will translate into customer visible instead of giving them a big clear red light or green light on release

Then an escape happens and all of a sudden it's about quality and not wanting to damage the company's reputation and we need to figure out how this happened

WhyteRyce fucked around with this message at 01:02 on Jun 26, 2020

SourKraut
Nov 20, 2005

POST QUALITY UNDER CONSTRUCTION


Geez guys, it's not like planes are falling out of the sky because of it.

priznat
Jul 7, 2009

Let's get drunk and kiss each other all night.

WhyteRyce posted:

Silicon validation is highly disrespected. People outside the industry think you just run benchmarks and game demos over and over again as a low level entry job. companies in the industry hate you because you're the guy holding up release and demand a huge amount of resources that can't be monetized or productized so it's just pure money down the drain to them. Program managers hate it because it either generates noise/false flags or finds weird corner cases they don't even know will translate into customer visible instead of giving them a big clear red light or green light on release

Then an escape happens and all of a sudden it's about quality and not wanting to damage the company's reputation and we need to figure out how this happened

Lol so goddamn accurate. The other problem is the amount of legacy poo poo that gets built up and no one can ever make the call to have a clean break or update things because it would require a resource investment for something that currently “works”.

CyberPingu
Sep 15, 2013

Ready To Ruck!





priznat posted:

It is hard won from sticking dozens of optane 16GB modules on aic adapters and those loving screws... Tiny stupid stubby screws!! Gah I hate them.



This is why god created the magnetic head screwdriver

SwissArmyDruid
Feb 14, 2014



Adventures with Das Frankenputer continues. Do I still need to microcode update or run a Ryzen-specific power plan, or is Windows aware enough that such things are un-necessary?

orcane posted:

According to that BIOS release list on their website they only dropped Bristol Ridge support when they added Ryzen 3000 (Zen 2) support, no mention of dropping Zen 1 CPUs so you should be able to just use the latest BIOS (but it won't really do much because there haven't been any Zen 1 improvements in AGESA for ages, but you'd be prepared for a hypothetical Ryzen 3950X ).

I'm still gonna take every motherboard stability improvement they deign to give me. Also, in case anyone wonders, they made it fit by ripping out the pretty mouse-driven UEFI GUI and replaced it with a much more spartan UEFI GUI.

HalloKitty
Sep 30, 2005

Adjust the bass and let the Alpine blast


SwissArmyDruid posted:

I'm still gonna take every motherboard stability improvement they deign to give me. Also, in case anyone wonders, they made it fit by ripping out the pretty mouse-driven UEFI GUI and replaced it with a much more spartan UEFI GUI.

MSI board? I think that nice greyscale BIOS is much nicer than the RGB mess with glowing buttons that comes as standard. I'd be happy if they were all like that

vv Argus monitor, great piece of software

HalloKitty fucked around with this message at 20:11 on Jun 27, 2020

90s Solo Cup
Feb 22, 2011

To understand the cup
He must become the cup





HalloKitty posted:

MSI board? I think that nice greyscale BIOS is much nicer than the RGB mess with glowing buttons that comes as standard. I'd be happy if they were all like that

Only thing I miss is the visual fan curve, which made setting fan speeds much easier.

spasticColon
Sep 22, 2004

In loving memory of Donald Pleasance

I got a Ryzen 3600 for 1080p60 gaming and I'm using the stock cooler but I've noticed that while running CPU intensive programs the CPU gets hot (up to 85 Celsius while running Cinebench; low to mid 70s Celsius while gaming) even with the CPU and case fans at full speed. Should I be worried? I looked at the specs for the 3600 on AMD's website and it says the max temp is 95C.

This is also my first AMD system since my old Athlon 64 X2 3800 system way back in 2005-2008 and if I remember correctly that chip ran kinda hot under load as well so do AMD chips still run hot?

Hasturtium
May 19, 2020

Netflix: HEY WE HAVE NEW SERIES
Me: I don't watch much series TV
Netflix: HERE'S AN AD FOR ONE SIX SECONDS AFTER THE LAST SHOW YOU JUST WATCHED
Me: No, man, quit it
Netflix: HOW ABOUT ANOTHER EDGY DOCUMENTARY
Me: Can I watch a movie from the seventies
Netflix: NO

spasticColon posted:

I got a Ryzen 3600 for 1080p60 gaming and I'm using the stock cooler but I've noticed that while running CPU intensive programs the CPU gets hot (up to 85 Celsius while running Cinebench; low to mid 70s Celsius while gaming) even with the CPU and case fans at full speed. Should I be worried? I looked at the specs for the 3600 on AMD's website and it says the max temp is 95C.

This is also my first AMD system since my old Athlon 64 X2 3800 system way back in 2005-2008 and if I remember correctly that chip ran kinda hot under load as well so do AMD chips still run hot?

The chiplet design tends to result in weird heat clustering that levels off over time, but the main issue is that the 3600’s stock cooler isn’t very good. You aren’t endangering the chip, but plopping a better cooler onto it would lower load temps and be quieter to boot.

Klyith
Aug 3, 2007

GBS Pledge Week


SwissArmyDruid posted:

Adventures with Das Frankenputer continues. Do I still need to microcode update or run a Ryzen-specific power plan, or is Windows aware enough that such things are un-necessary?

You don't need to run a Ryzen-specific power plan, windows is now aware of how ryzen works and the defaults are fine. You get the Ryzen Balanced plan included when you install the chipset drivers, use it if you want. Even the 1ismus is just some small differences that are no more than a 1% change up or down in resulting performance.

spasticColon posted:

I got a Ryzen 3600 for 1080p60 gaming and I'm using the stock cooler but I've noticed that while running CPU intensive programs the CPU gets hot (up to 85 Celsius while running Cinebench; low to mid 70s Celsius while gaming) even with the CPU and case fans at full speed. Should I be worried? I looked at the specs for the 3600 on AMD's website and it says the max temp is 95C.

This is also my first AMD system since my old Athlon 64 X2 3800 system way back in 2005-2008 and if I remember correctly that chip ran kinda hot under load as well so do AMD chips still run hot?

That's normal for the stock cooler. They don't thermal throttle until 95, but that's still safe -- the CPU is faster at cutting power than the heat is at damaging anything.

And no, AMD chips are actually cool compared to intel these days! Intel raised their TJmax to 100C because they're so hot. Modern silicon can get pretty drat hot before it hurts itself.

kimcicle
Feb 23, 2003



I finished a Ryzen 3600 + ASUS ROG B550-F build but there's something funky going on with my system. At system idle, the CPU temp will jump around 10 degrees every few seconds. This causes the system fans to ramp up, but the temp settles back towards it's normal idle temp then spikes again. On task manager my GPU speed keeps fluctuating, hovering around the 3.9 / 4.0 GHz range. Any ideas as to where I can look, or is this something I'll have to plan my fan speeds around?

Fwoderwick
Jul 14, 2004



Hasturtium posted:

The chiplet design tends to result in weird heat clustering that levels off over time, but the main issue is that the 3600’s stock cooler isn’t very good. You aren’t endangering the chip, but plopping a better cooler onto it would lower load temps and be quieter to boot.

I'm currently planning to build a PC with a 3600 and this 'stock cooler isn't very good' sentiment seems to be a universal constant when discussing it. I was a bit oblivious to the concept of CPU coolers before this build so trying to get my head around what is/isn't necessary. I'm keen on keeping noise down under load and not having temperature hamstring it's effectiveness but I'm decidedly mid-range when it comes to PC parts.

In the PC Builder thread sean10mm's put up a suggested build for a 3700x with a Noctua NH-U14S (~$65/£60). Does that sound like a solid choice? I've seen a few mentions of adding coolers that are half that price but I'm a bit skeptical that something which is <$30/£30 is going to be all that good.

And on a tangent is the stock cooler with the 3700x any good or is replacing that cooler recommended too? If I got something like the Noctua I'm already 2/3rds of the way through the price difference to that CPU...

Klyith
Aug 3, 2007

GBS Pledge Week


kimcicle posted:

I finished a Ryzen 3600 + ASUS ROG B550-F build but there's something funky going on with my system. At system idle, the CPU temp will jump around 10 degrees every few seconds. This causes the system fans to ramp up, but the temp settles back towards it's normal idle temp then spikes again. On task manager my GPU speed keeps fluctuating, hovering around the 3.9 / 4.0 GHz range. Any ideas as to where I can look, or is this something I'll have to plan my fan speeds around?

Zen 2 desktop ryzens boost to max clock like an over-excited puppy, even when the system is idle. The boost is for a very short amount of time before realizing that no, there's no real work to do... but since they control the fan based on the hottest core temp that makes them spin up the fan.

There are two ways to manage this with fan speed controls:
1. raise the low-temp fan speeds so that the cooler is working harder even at idle and absorbs the heat spikes better, but it's louder
2. push the ramp to high fan speed to higher temps such that the +10C momentary boosts don't trigger fan ramp, but you lose a hair of performance since it's hotter
3. get a cheap 120mm tower cooler

Fwoderwick posted:

I'm currently planning to build a PC with a 3600 and this 'stock cooler isn't very good' sentiment seems to be a universal constant when discussing it. I was a bit oblivious to the concept of CPU coolers before this build so trying to get my head around what is/isn't necessary. I'm keen on keeping noise down under load and not having temperature hamstring it's effectiveness but I'm decidedly mid-range when it comes to PC parts.

In the PC Builder thread sean10mm's put up a suggested build for a 3700x with a Noctua NH-U14S (~$65/£60). Does that sound like a solid choice? I've seen a few mentions of adding coolers that are half that price but I'm a bit skeptical that something which is <$30/£30 is going to be all that good.

The stock cooler is adequate because it keeps the CPU from throttling, you will not lose any performance from it. It's not very good from a perspective of noise. Now keep in mind that many intel stock coolers don't even accomplish the throttling part. So from that standpoint the stock AMD sinks are pretty good. But stock sinks are cheap and the way they engineer to cost is to have 2k RPM fans that are loud.


A cheap $30 120mm cooler is fine for a 3600. My experience with an arctic 33 (on a 1600X, about the same TDP-wise) is that it never needed more than 1200rpm for prime95, and in games generally was just a bit above 1000rpm. Not whisper-quiet but not at all annoying. They don't have the idle-boost problem that the stock cooler has, because heat pipes are good at absorbing spikes. And you can turn on PBO to get some OC if you want.

OTOH the noctua U14S on a 3600 or 3700X will be pretty quiet even when the CPU is running flat out & OCed. That's what you get for twice the money. If silence is a priority it's a good investment. The noctua U14S gets suggested with the 3700X because if you've picked a 3700X rather than a 3600 that suggests you have some discretionary budget for luxuries.

Fwoderwick posted:

And on a tangent is the stock cooler with the 3700x any good or is replacing that cooler recommended too? If I got something like the Noctua I'm already 2/3rds of the way through the price difference to that CPU...

The stock cooler that comes with a 3700X is, well, it would be well paired for a 3600. On the 3700X, which has 2 more cores and is hotter than a 3600, it's still loud under load. It has heatpipes, which mean it can deal with transient spikes much better. I would still want a real heatsink.

Fwoderwick
Jul 14, 2004



Awesome, thanks for the detailed response and lower priced suggestion.

I'm still weighing up what total cost I'm happy with as I'm a miser at heart but will spend more for a quality mid-range product if it's a good investment. I like what I've read about the Noctua performance and aftercare, but will read up about the Arctic in case I'm just getting caught up in the excitement of it all and overspeccing for my needs.

future ghost
Dec 5, 2005

det er noget at leve for

Gun Saliva

Klyith posted:

That's normal for the stock cooler. They don't thermal throttle until 95, but that's still safe -- the CPU is faster at cutting power than the heat is at damaging anything.

And no, AMD chips are actually cool compared to intel these days! Intel raised their TJmax to 100C because they're so hot. Modern silicon can get pretty drat hot before it hurts itself.
Intel actually allowed raising it to 110C with the 9-series since anything using enough AVX can push 100C on the 9700k/9900k even at stock.

kimcicle
Feb 23, 2003



Klyith posted:

Zen 2 desktop ryzens boost to max clock like an over-excited puppy, even when the system is idle. The boost is for a very short amount of time before realizing that no, there's no real work to do... but since they control the fan based on the hottest core temp that makes them spin up the fan.

There are two ways to manage this with fan speed controls:
1. raise the low-temp fan speeds so that the cooler is working harder even at idle and absorbs the heat spikes better, but it's louder
2. push the ramp to high fan speed to higher temps such that the +10C momentary boosts don't trigger fan ramp, but you lose a hair of performance since it's hotter
3. get a cheap 120mm tower cooler

Thanks for the reply. I'm not using the stock cooler, and it looks like only the chassis fans were spinning up so I'll mess with the fan curves some more.

gradenko_2000
Oct 5, 2010



Lipstick Apathy

I'm irrationally angry that AMD CPUs did not have AVX2 until Ryzen, because that's two old computers I've built that now can't use virtual greenscreens in Zoom/Teams

Adbot
ADBOT LOVES YOU

Statutory Ape
Sep 12, 2017



my cpu makes me feel othered

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply
«432 »