|
BurritoJustice posted:The changes are cool and good because even though they're still arbitrary they make it easier to compare to other fabs. if I'm reading this right: TSMC's 7nm, which is what's used for AMD's Zen 2/3, has a transistor density of 91.20 Intel's 10nm (SuperFin, whatever), which is what's used for Tiger Lake, has a transistor density of 100.76 They're renaming 10nm to "Intel 7", which is going to sound like "7 nm", which is appropriate because the transistor density is similar?
|
#
?
Jul 27, 2021 09:35
|
|
|
|
| # ? May 8, 2024 05:35 |
|
|
i get why they don't do it but if they just labelled by transistor density it would make so much more sense
|
|
#
?
Jul 27, 2021 09:41
|
|
|
BurritoJustice posted:You see enough people taking the numbers literally and thinking TSMC 7nm is a generation ahead of Intel 10nm ESF when it's mostly on par. So you're saying the naming was correct and they shouldn't have changed it? 
|
|
#
?
Jul 27, 2021 11:27
|
|
|
Shipon posted:i get why they don't do it but if they just labelled by transistor density it would make so much more sense "Pfft, TSMC 91.20 MTr/mm2 has nothing on Intel 100.76 MTr/mm2, AMD is screwed this gen"
|
|
#
?
Jul 27, 2021 12:24
|
|
|
Tbh, it's kinda weird that manufacturing process became this huge marketing thing anyways so as far egregious tech naming schemes go in my opinion this doesn't move the needle. Intel has historically been more conservative with their process names as noted above, probably in part because they're their own biggest customer.
|
|
#
?
Jul 27, 2021 13:01
|
|
|
Shipon posted:i get why they don't do it but if they just labelled by transistor density it would make so much more sense would require everyone's brain shifting from smaller number = better to bigger number = better, which is why i bet they don't do it. BurritoJustice posted:"Pfft, TSMC 91.20 MTr/mm2 has nothing on Intel 100.76 MTr/mm2, AMD is screwed this gen" you'd just call it like 101 or 45 or whatever
|
|
#
?
Jul 27, 2021 13:31
|
|
|
VorpalFish posted:Tbh, it's kinda weird that manufacturing process became this huge marketing thing anyways so as far egregious tech naming schemes go in my opinion this doesn't move the needle. twisting the truth for $$$? well i never
|
|
#
?
Jul 27, 2021 16:16
|
|
|
gradenko_2000 posted:if I'm reading this right: Intel PR7+
|
|
#
?
Jul 27, 2021 16:39
|
|
|
BurritoJustice posted:The changes are cool and good because even though they're still arbitrary they make it easier to compare to other fabs. Just go with density, i like this chart.
|
|
#
?
Jul 27, 2021 16:59
|
|
|
But peak theoretical density is kinda meaningless too! And there isn’t a standard for measuring the density anyway! Oh no!
|
|
#
?
Jul 27, 2021 17:09
|
|
|
Reminds me of the old Athlon naming scheme, putting an "Intel Equivalent" mhz speed in the model name
|
|
#
?
Jul 27, 2021 17:16
|
|
|
Wasnt that “equivalent to 2500 mhz versus an amd 1000 mhz”
|
|
#
?
Jul 27, 2021 17:17
|
|
|
Even though it was obviously for intel
|
|
#
?
Jul 27, 2021 17:17
|
|
|
they should bring that back imo, just brand every cpu relative to the pentium 4 where's my Ryzen 50,000+
|
|
#
?
Jul 27, 2021 17:19
|
|
|
They still do that as far as Ryzen 3, 5, 7, and 9 being positioned against an i3, i5, i7, and i9
|
|
#
?
Jul 27, 2021 17:20
|
|
|
Wild EEPROM posted:Wasnt that “equivalent to 2500 mhz versus an amd 1000 mhz” The 2500+ ran at 1.8ghz stock and a lot more overclocked. Barton was legit, marketing aside.
|
|
#
?
Jul 27, 2021 21:00
|
|
|
We just label everything by Linux kernel compile time, single thread and multithread.
|
|
#
?
Jul 28, 2021 00:53
|
|
|
That's basically what we do now with cinebench!
|
|
#
?
Jul 28, 2021 02:29
|
|
|
gradenko_2000 posted:They still do that as far as Ryzen 3, 5, 7, and 9 being positioned against an i3, i5, i7, and i9 There's also the hilariously petty chipset naming schemes that have been going back and forth. AMD starting their HEDT line with X399 when Intel was X299, and now the back and forth with 370/390/570/590 etc
|
|
#
?
Jul 28, 2021 07:10
|
|
|
it's not petty it's just dumb
|
|
#
?
Jul 28, 2021 07:38
|
|
|
Putting aside the node renaming stuff, the idea of delivering power to transistors from one side so the data lines have the other side to themselves seems like a great innovation.
|
|
#
?
Jul 28, 2021 08:41
|
|
|
https://videocardz.com/newz/intel-only-to-launch-alder-lake-kf-series-and-z690-chipset-this-year-the-rest-coming-at-ces-2022 Intel is giving out incentives for motherboard makers to use ATX12VO apparently, they really want it to be a thing
|
|
#
?
Jul 28, 2021 14:58
|
|
|
Maybe related to California's recent idle power efficiency phase in? I know people were speculating that's why they even made 12vo to begin with. Internet currently appears to be freaking out over Dell not selling most alienware models in a bunch of states due to the phase in.
|
|
#
?
Jul 28, 2021 15:13
|
|
|
repiv posted:https://videocardz.com/newz/intel-only-to-launch-alder-lake-kf-series-and-z690-chipset-this-year-the-rest-coming-at-ces-2022 OEMs like Dell seem to be trying very hard to make it happen so Intel probably only wants to deal with that.
|
|
#
?
Jul 28, 2021 15:18
|
|
|
repiv posted:https://videocardz.com/newz/intel-only-to-launch-alder-lake-kf-series-and-z690-chipset-this-year-the-rest-coming-at-ces-2022 ATX12vo wouldn't be a bad thing to go to it would help idle power consumption a bunch
|
|
#
?
Jul 28, 2021 15:57
|
|
|
I'm all for motherboards being mechanically and electrically simplified, and very much in favor of getting rid of the 2-part 24-pin power connector and it's huge wad of wires. Please bring on the single 12V standard (and force AMD to use it too).
|
|
#
?
Jul 28, 2021 16:54
|
|
|
In principle I think the new 12V is fine. But I feel like they wasted the chance to redo all the power delivery to the board. What I saw from GamersNexus looked like you still needed to run a separate 4 or 8 pin 12V cable for the CPU instead of just having one big block. But this was one of those terrible OEM systems so it can probably be done better.
|
|
#
?
Jul 28, 2021 17:09
|
|
|
tbh i think the biggest issue is going to be people using cheap $1 adapters (which are going to be plentiful) and then burning down their house
|
|
#
?
Jul 28, 2021 17:24
|
|
|
Icept posted:In principle I think the new 12V is fine. But I feel like they wasted the chance to redo all the power delivery to the board. What I saw from GamersNexus looked like you still needed to run a separate 4 or 8 pin 12V cable for the CPU instead of just having one big block. But this was one of those terrible OEM systems so it can probably be done better. Is that this article? https://www.gamersnexus.net/guides/3568-intel-atx-12vo-spec-explained-what-manufacturers-think The big surprise for me there is this quote: quote:We also asked Gerow if economies of scale and mass production would move ATX12VO and ATX12V power supply standards toward each other, eventually getting into the DIY space. He replied: "Not really. Dell, HP and Lenovo already use a 12V Only type of solution, but their connectors are proprietary. Intel is just taking that idea and trying to standardize it. I put together a new PC every 3-5 years, and never reuse power supplies, so the complaint about backwards compatibility doesn't really resonate with me. So, besides the backwards compatibility thing, what's the bothersome part for DIY/enthusiasts?
|
|
#
?
Jul 28, 2021 17:58
|
|
|
Kazinsal posted:Yeah, except it's the former "Intel 4nm", so it should really be 40Å, since 1nm == 10Å. They should call it 40A, because that is how many amps it's going to draw at 12V
|
|
#
?
Jul 28, 2021 18:04
|
|
|
canyoneer posted:I put together a new PC every 3-5 years, and never reuse power supplies, so the complaint about backwards compatibility doesn't really resonate with me. So, besides the backwards compatibility thing, what's the bothersome part for DIY/enthusiasts? The change doesn't really benefit the enthusiasts, so it is a bit of a net negative for DIY gaming/workstation folks specifically. The efficiency gains are negligible to non existent for a high power gaming computer (it is really just designed for bulk OEM to meet bulk standards), it is going to increase motherboard costs and complexity likely without any drop in power supply costs, it may make the entire compatibility situation much more complicated for those that still have 3.3v/5v devices ("oh no, this cheap motherboard only has 1 sata power plug!"), you are leaving the 3.3v/5v regulation in the hands of 5-6 companies that basically all have massively hosed up voltage circuitry selection at one point or other on their current products, and it is sorta halfassed for enthusiasts in that it didnt actually consolidate the number or ease of use of the cables. In the end, I don't actually think it will be that bad because I think the DIY space is just going to ignore the standard for years to come. But it will likely be annoying for all of us who do computer janitor duties for friends and family.
|
|
#
?
Jul 28, 2021 18:52
|
|
|
mdxi posted:I'm all for motherboards being mechanically and electrically simplified, and very much in favor of getting rid of the 2-part 24-pin power connector and it's huge wad of wires. Please bring on the single 12V standard (and force AMD to use it too). Wait, so the motherboards are getting simpler? I heard some complaining earlier about how all the voltage conversion was getting pushed down onto motherboard OEMs to get 3.3V and other rails
|
|
#
?
Jul 28, 2021 21:07
|
|
|
I suppose putting the 3.3v/5v converters right next to where they're needed might simplify things compared to having to route three different voltages across the entire board? If it lets them use fewer PCB layers it could end up being cheaper despite having more components.
|
|
#
?
Jul 28, 2021 21:10
|
|
|
Nomyth posted:Wait, so the motherboards are getting simpler? I heard some complaining earlier about how all the voltage conversion was getting pushed down onto motherboard OEMs to get 3.3V and other rails You can view it either way I suppose, but to be honest I was thinking slightly out of context anyway. My experience with single-voltage mobos is in large datacenters, where everything that a machine doesn't need gets thrown off the board, so it's physically simpler. But of course, now that I'm thinking about it, that won't be true of DIY mobos. There is a lot of stuff on DIY mobos that I'd be happy to ditch, but that's completely subjective.
|
|
#
?
Jul 28, 2021 22:03
|
|
|
canyoneer posted:Is that this article? re: big surprise, 12VO is a very very old idea that has shipped in many real systems. Every single Apple Mac Pro ever built has a 12VO power supply, and that started in 2006 with the MacPro1,1 model (a dual socket Woodcrest workstation). Think about it this way: ATX is a 1995 standard, which itself was just an attempt to codify and standardize what people were doing when cloning and extending the 1980s IBM PC AT. It's incredibly dated at this point. Everyone with the freedom to not be tied down to that crusty ad hoc retrofit standard has built systems which move beyond it. It's kinda weird that Intel hasn't tried to push something like ATX12VO long before now. TLDR: 12VO is long overdue.
|
|
#
?
Jul 28, 2021 23:22
|
|
|
Bring back BTX and modernize it, so pcie cards primary side faces up and has standardized cooling etc. BTX+ baby
|
|
#
?
Jul 28, 2021 23:28
|
|
|
hobbesmaster posted:OEMs like Dell seem to be trying very hard to make it happen so Intel probably only wants to deal with that. The weird Dell power supply/motherboard seen in thier PC reviewed by GamersNexus didn't actually use the ATX12VO standard. It used their own proprietary standard that doesn't work with anything else.
|
|
#
?
Jul 29, 2021 01:41
|
|
|
Cygni posted:The change doesn't really benefit the enthusiasts, so it is a bit of a net negative for DIY gaming/workstation folks specifically. Smaller PSUs and less annoying cabling sounds great for anyone building an itx system tbf.
|
|
#
?
Jul 30, 2021 09:19
|
|
|
Arzachel posted:Smaller PSUs and less annoying cabling sounds great for anyone building an itx system tbf. Ah, but the motherboard has to do the 3.3V and 5V conversion now, and cant ditch em cause of SATA, USB, fan controllers, and aRGB (seriously)... so the question is where exactly do they stick that, including the SATA power connectors, without losing features on ITX boards that are already stacking M.2 drives vertically to save space? ITX is one of the reasons I don't see 12VO taking off in the DIY space for a while. Its possible that the SFF market goes the way of those little DC to DC pico PSUs with a 12V passthrough i guess? I dunno. I think its more likely they just keep making regular 24pin PSUs and start shipping with a converter (or separate modular cable) if you are running a 12VO motherboard instead. Just as an example, here is what that section looks like on one of the few 12VO DIY boards out there: 
Cygni fucked around with this message at 18:56 on Jul 30, 2021 |
|
#
?
Jul 30, 2021 18:54
|
|
|
|
| # ? May 8, 2024 05:35 |
|
|
ITX motherboards could switch to SO-DIMMs and hide those on the underside like they already do to M.2 drives.
|
|
#
?
Jul 31, 2021 01:02
|
|