|
https://www.theverge.com/2021/7/29/22599092/intel-nuc-11-extreme-beast-canyon-specs-price-availability-preview Good idea, glad they continue to work on it, they've upgraded to a fan hat and a better dust filter grill, but Intel refuses to do the needful thing and flip the orientation of the Compute Element cartridge so it can pull air in from the opposite side and use a better cooler than a laptop blower.
|
# ? Jul 31, 2021 01:15 |
|
|
# ? May 9, 2024 00:41 |
|
wargames posted:ATX12vo wouldn't be a bad thing to go to it would help idle power consumption a bunch does it really? saving 10% power at idle doesn't seem worth having to move all the power conversion to the motherboard I just don't want to have to dump the 850W seasonic PSU i bought a few years ago that's got a solid 10 year warranty because of some negligible savings in idle usage
|
# ? Jul 31, 2021 01:15 |
|
I don't see why you'd have to, current PSUs already do the 12V part of 12VO
|
# ? Jul 31, 2021 02:09 |
|
Shipon posted:does it really? saving 10% power at idle doesn't seem worth having to move all the power conversion to the motherboard https://www.guru3d.com/news-story/atx12vo-power-supply-standard-gets-testedshows-really-good-idle-power-consumption,2.html This its better then 10% because the idle here is 7 watts vs like 15-25w with the old system?
|
# ? Jul 31, 2021 04:24 |
|
Shipon posted:does it really? saving 10% power at idle doesn't seem worth having to move all the power conversion to the motherboard https://www.youtube.com/watch?v=heyGtgdfN7A It is closer to 50% at idle in ideal conditions. I'd imagine there could be a market for specialist SATA-less 12VO motherboards for particular small form factor builds. Going all m.2 is already quite popular, with network attached mass storage.
|
# ? Jul 31, 2021 04:27 |
|
The United States posted:https://www.theverge.com/2021/7/29/22599092/intel-nuc-11-extreme-beast-canyon-specs-price-availability-preview They're still doing that stupid poo poo where the CPU element's intake is restricted by a GPU sandwich preheating the intake air, instead of putting all the componentry and cooling on the OPPOSITE side of the board, much as how such SFF cases as the NCase or DANCase achieve with a riser cable. But those cases have to operate within the constraints of standardized off-the-shelf parts, while Intel is already doing weird-rear end poo poo, and should really just go full ASRock on this, instead of this bullshit half-step. SwissArmyDruid fucked around with this message at 05:32 on Jul 31, 2021 |
# ? Jul 31, 2021 05:04 |
|
what the gently caress with the face? it's a psu bro calm down.
|
# ? Jul 31, 2021 06:01 |
|
BurritoJustice posted:https://www.youtube.com/watch?v=heyGtgdfN7A I've got news you are going to love: USB requires 5V, and M.2 uses 3.3V. So SFF is gonna be a mess. (or 12VO will just be ignored, which is likely)
|
# ? Jul 31, 2021 06:24 |
|
Cygni posted:I've got news you are going to love: USB requires 5V, and M.2 uses 3.3V. So SFF is gonna be a mess. (or 12VO will just be ignored, which is likely) Yeah, I was more thinking the weird clunky extra power connectors on the board. Should minimise the extra PCB space if they don't need to worry about anything external. B-1.1.7 Bomber posted:what the gently caress with the face? it's a psu bro calm down. It worked, he pulled 1.1M views on a short video about a mundane power supply standard by pulling a whacky face and using an ambiguous title
|
# ? Jul 31, 2021 12:02 |
|
1.1M on the low end for him. He clearly needed to do an even more absurd face.
|
# ? Jul 31, 2021 12:11 |
|
Cygni posted:I've got news you are going to love: USB requires 5V, and M.2 uses 3.3V. So SFF is gonna be a mess. (or 12VO will just be ignored, which is likely) I feel like a couple buck converters just aren't that big of a deal. Now whether there's anyone besides me who would buy a board with no sata is a different story
|
# ? Jul 31, 2021 13:15 |
|
ATX12VO is a phenomenally bad idea simply because I don't trust mobo makers to not go stupidly cheap on the voltage conversion circuitry on their low-end to mid-range boards. OR to go unnecessarily complex on their high-end boards just so they can make $700+ boards "the norm" from now on, because "we already do that on our video cards and we know people will 'just buy it' now." It's not going to save anyone any money. What "cost savings" might be applied to PSUs will be lost from need of retooling on their end, and mobo makers will just charge more for having to put new poo poo on their boards. BIG HEADLINE fucked around with this message at 14:08 on Jul 31, 2021 |
# ? Jul 31, 2021 13:35 |
|
wargames posted:https://www.guru3d.com/news-story/atx12vo-power-supply-standard-gets-testedshows-really-good-idle-power-consumption,2.html
|
# ? Jul 31, 2021 13:48 |
|
BIG HEADLINE posted:ATX12VO is a phenomenally bad idea simply because I don't trust mobo makers to not go stupidly cheap on the voltage conversion circuitry on their low-end to mid-range boards. OR to go unnecessarily complex on their high-end boards just so they can make $700+ boards "the norm" from now on, because "we already do that on our video cards and we know people will 'just buy it' now." i'm with this guy
|
# ? Jul 31, 2021 15:44 |
|
Cygni posted:I've got news you are going to love: USB requires 5V, and M.2 uses 3.3V. So SFF is gonna be a mess. (or 12VO will just be ignored, which is likely) Here's a DC-DC buck converter module which I found in a couple minutes browsing TI's site: https://www.ti.com/document-viewer/TPSM5D1806/datasheet/GUID-249ADD81-B7B6-48B9-B1E7-80F2C438E743#TITLE-SLUSDV8SLVSE269359 4.5V to 15V input, dual 6A outputs, 0.5V to 5.5V output range (60W total at 5V) in an 8mmx5.5mm QFN package. Granted, being optimized for cost, most PC motherboards will use discrete components rather than fancy modules. I'm just using this component's existence to point out that SFF should be fine: you can pack a lot of buck converter in a very tiny volume. BIG HEADLINE posted:ATX12VO is a phenomenally bad idea simply because I don't trust mobo makers to not go stupidly cheap on the voltage conversion circuitry on their low-end to mid-range boards. OR to go unnecessarily complex on their high-end boards just so they can make $700+ boards "the norm" from now on, because "we already do that on our video cards and we know people will 'just buy it' now." Sure, on one level, the usual suspects aren't great. If you want something actually high quality you have few or no options. But pull it back a bit. Providing Vcore is also a DC-to-DC conversion, but CPU Vcore requires fantastically precise voltage regulation, digitally programmable variable voltage output with very fine increments and high slew rate requirements (which don't permit much over/undershoot at the end of the transition), >100A current output, and a ton of other very difficult things. By comparison, fixed 5V +/- 5% at a few amps is a walk in the park. If you suspect the mobo maker you're considering patronizing can't pull it off, why the gently caress would you have ever thought them capable of providing good Vcore? I'll also point this out: Every five dollar USB thumbdrive has a buck converter inside, because lol no the chip(s) in them can't run direct from 5V. Same applies to M.2 drives and the 3.3V supply there. Cheap, efficient, and reliable point-of-load buck regulators are a solved problem and you rely on them all the time without knowing it.
|
# ? Jul 31, 2021 19:57 |
|
BIG HEADLINE posted:ATX12VO is a phenomenally bad idea simply because I don't trust mobo makers to not go stupidly cheap on the voltage conversion circuitry on their low-end to mid-range boards. OR to go unnecessarily complex on their high-end boards just so they can make $700+ boards "the norm" from now on, because "we already do that on our video cards and we know people will 'just buy it' now." I agree with this
|
# ? Jul 31, 2021 22:10 |
|
If it saves me the bulk of the cabling that the current 24-pin cable package has, I'm all for it. As has been pointed out, 12v->5v/3v is neither difficult nor expensive, and there's really no reason to believe that motherboards won't be able to figure that one out at a trivial cost. I mean, gently caress guys, we're talking about motherboard manufacturers who are having to figure out how the hell to manage the power delivery and voltage conversions for DDR5 and a CPU. Figuring out how to throw 5v for your USB drives on there is basically an afterthought by comparison.
|
# ? Jul 31, 2021 22:28 |
|
Not only is it not hard, it's not rare. There are already many perfectly sufficient off the shelf solutions. It's definitely the way forward, it's just a question of when it happens.
|
# ? Jul 31, 2021 22:34 |
|
Real questions: when does USB-c or its successor add a 12V only spec. USB-PD already allows 12V, I believe.
|
# ? Aug 1, 2021 01:23 |
|
K8.0 posted:Not only is it not hard, it's not rare. There are already many perfectly sufficient off the shelf solutions. It's definitely the way forward, it's just a question of when it happens. No its not. Why not have the actual power supply supply .. power.
|
# ? Aug 1, 2021 01:26 |
|
redeyes posted:No its not. Why not have the actual power supply supply .. power. Its easier to deliver more power at 12V than at 3.3V or 5V. I^2R after all. I think USB-C, 3.2 and USB-PD are really screwing things up, motherboards are not prepared to deliver 15W at 5V potentially to multiple devices let alone the full USB PD spec which has the options of 5/9/15/20V at up to 100W. If motherboards are actually doing to support that, more copper spent on 12V the better. Really though at that point you'd probably be thinking about 48VDC or something.
|
# ? Aug 1, 2021 01:39 |
|
redeyes posted:No its not. Why not have the actual power supply supply .. power. Because 24 fat-rear end cable pins is obnoxious as gently caress, and because it lets the entire system run more efficiently.
|
# ? Aug 1, 2021 01:49 |
|
redeyes posted:No its not. Why not have the actual power supply supply .. power. Why not have the power supply supply vcore and every other voltage? We need a 900 pin connector!
|
# ? Aug 1, 2021 02:04 |
|
K8.0 posted:Why not have the power supply supply vcore and every other voltage? We need a 900 pin connector! Alienware used *this* monstrosity once: Thankfully the rest of the system was standard ATX. That connector linked up with the PSU.
|
# ? Aug 1, 2021 02:14 |
|
BIG HEADLINE posted:Alienware used *this* monstrosity once: What was the point of this?
|
# ? Aug 1, 2021 02:21 |
|
Shrimp or Shrimps posted:What was the point of this? Nothing inside modern Dell computers makes any sense.
|
# ? Aug 1, 2021 02:58 |
|
BIG HEADLINE posted:Alienware used *this* monstrosity once: I'm not seeing Loss no matter how hard I try.
|
# ? Aug 1, 2021 07:24 |
|
If Intel wants to push ATX12VO then Intel should include the 12-to-5s and 12-to-3.3s in their chipset. gently caress the shelf
|
# ? Aug 1, 2021 08:06 |
|
Some hope? https://www.pcgamer.com/uk/amp/intel-atx12vo-power-standard-pushback-manufacturers/
|
# ? Aug 1, 2021 13:39 |
|
WattsvilleBlues posted:Some hope? im the writer who thinks classic ATX uses AC power quote:The proposed power standard would deliver just a single rail of 12V direct current through a 10-pin connector—as opposed to the current 24-pin, alternating current standard—but would require a complete rehash of motherboard and component designs.
|
# ? Aug 1, 2021 13:46 |
|
I imagine that ATX12VO might get some purchase in the OEM space, where manufacturers might be okay with a redesign of boards if it's for an order in the tens or hundreds of thousands of machines in one go this would also be the kind of thing where cutting idle power usage yields benefits at scale. If you go from 15 watts to 7 watts across the 1,000 machines on a call center, that's not nothing.
|
# ? Aug 1, 2021 13:46 |
|
gradenko_2000 posted:this would also be the kind of thing where cutting idle power usage yields benefits at scale. If you go from 15 watts to 7 watts across the 1,000 machines on a call center, that's not nothing. Yeah, this is almost the entire point of energy efficiency regulations in consumer products. It really seems like a no-brainer to me to change the PC power supply standards to improve power efficiency. I seriously doubt that the the old PC power supply standards put PCs close to the technological limit, and that changing the standard would compromise other parts of the system.
|
# ? Aug 1, 2021 19:22 |
|
Shipon posted:does it really? saving 10% power at idle doesn't seem worth having to move all the power conversion to the motherboard Corsair already appears to have a 12vo adapter cable for their existing modular psus and I have a hard time believing seasonic wouldn't do the same if 12vo becomes big. Seems like using a 12vo motherboard with an existing serviceable atx psu would be an extremely solveable problem.
|
# ? Aug 1, 2021 20:55 |
|
People hang on to PSUs too long anyway and it's not like the transition will happen overnight. Once it's happening if you buy a PSU that isn't 12VO compatible and want to use it for your next build, the dingus is you and not anyone else.
|
# ? Aug 1, 2021 21:24 |
|
I can see wanting to continue using a good quality psu that's inside waranty period instead of buying a new one when upgrading with a 12vo motherboard, and overall that's better for the environment. But again, seems like a problem with a very simple solution.
|
# ? Aug 1, 2021 21:46 |
|
I’m not worried about it because (and i could be completely wrong about any or every part of this) - the circuitry and components aren’t expensive or rare - if dell and hp, companies well known for pinching out cheap pc’s by the million can do it, then theres no reason why companies like asusor msi can’t. - those micro desktops for offices already do it with a power brick and those bricks are absolute garbage tier quality - adapters already exist / will exist - most good psus are already capable of putting out their whole output on the 12v so its not like the old days of cross loads - seasonic made a psu which only does 12v and has an additional module which does the 5 and 3.3 and the 24 pin comes out of that. With just a new set of cables and not using that module, its basicallt 12vo - the enthusiast builders and small companies are a tiny part of the market
|
# ? Aug 2, 2021 01:19 |
|
I think its a great idea, the main reason I don't repurpose gaming rigs into servers is because the idle power is too much, it has to be <15W for me to run 24/7.
Perplx fucked around with this message at 01:55 on Aug 2, 2021 |
# ? Aug 2, 2021 01:49 |
|
K8.0 posted:People hang on to PSUs too long anyway Yeah, it sucks so much that I haven't needed to buy a new PSU for the last 15 years!
|
# ? Aug 2, 2021 03:55 |
|
Perplx posted:I think its a great idea, the main reason I don't repurpose gaming rigs into servers is because the idle power is too much, it has to be <15W for me to run 24/7. That's not what "idle" power means. Idle power is when it's off, just wasting energy waiting on someone to press the power button. Minimal current to keep the USB (and possibly ethernet) circuitry on for wakeup events. You're still looking at near 3 digit wattage on a server "not doing anything" but waiting for a connection. Mine never goes below 120. ~Coxy posted:
Harik fucked around with this message at 09:55 on Aug 2, 2021 |
# ? Aug 2, 2021 09:52 |
|
|
# ? May 9, 2024 00:41 |
|
E: gently caress, Q!=E, haven't done that in a long time.
|
# ? Aug 2, 2021 09:55 |