Jump to content

moriz1

Member
  • Posts

    18
  • Joined

  • Last visited

Awards

This user doesn't have any awards

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

moriz1's Achievements

  1. i expect the cardboard will be replaced by something more suitable in the final build. as for 1U and full-height motherboards: you're wrong. 1U server cases are typically designed to fit the various ATX/mATX/mITX sizes. whether THIS particular case correctly follows the spec is anyone's guess, but since this is more of a one-off show piece, i expect that it doesn't follow it too closely, hence the issue with standoffs and back insulation. and no matter what, 1U server cases aren't designed to hold laptop parts.
  2. what's the point in using laptop parts, when they can already use standard desktop parts? there's no benefit, both in terms of cost, performance, and reparability.
  3. A lot of anticheat software are specifically capable of detecting whether if you are running the game in a VM. There are ways around them, but it becomes a bit of an arms-race as each side tries to beat the other.
  4. qemu has a built in virtual switch that all VMs are automatically connected to. so, networking isn't an issue.
  5. ssh into your proxmox server, and enter: nano /etc/pve/qemu-server/<vmid>.conf where <vmid> is the id of your VM make sure you have these lines in the conf file: args: -cpu 'host,+kvm_pv_unhalt,+kvm_pv_eoi,hv_vendor_id=NV43FIX,kvm=off,-hypervisor cpu: host,hidden=1,flags=+pcid there are probably a few flags in there that are unnecessary. but specifically, you need the -hypervisor flag in args, and host,hidden=1 in cpu. if in doubt, just use all of them. also, note that doing this will likely result in a small loss of performance, since it also turns off a few optimizations for virtualized CPUs. i know for a fact that this gets around the anticheat for Genshin Impact (and basically all Hoyoverse games). i am not sure if it gets around other anticheats like Battleye, EAC etc.
  6. they were pretty clear that this video is about the proof of concept, not a final build. so yes, there will be another video (probably) where they put it together for deployment. realistically speaking, if they want the system to be even vaguely compact (4U or smaller), they have to go with water cooling. might even be a good opportunity to make that rackmount gaming case that Linus was talking about in a few WAN shows a few weeks ago.
  7. i run something similar to this, but on a much smaller scale. i have a number of GPU VMs that could run games, and my friends connect to them using parsec. i also use one VM to play games when i'm visiting my parents, just so i don't have to lug around a heavy gaming laptop. unlike Linus, i'm using Proxmox. one thing Linus and co needs to watch out for, is that they have to figure out a way to hide the fact that the VMs are... VMs. Windows is capable of detecting whether it is running as a VM or not, and some anticheats can also do the same. there are a number of CPU flags that i have to use (some of them hidden from the GUI, so i had to edit the config files manually to enable) to accomplish this. prior to doing so, some games, like Genshin Impact, would refuse to run. a quick way to check, is to open up task manager and click on the CPU tab in the Performance tab. If you see the line "Virtual machine: Yes", and the CPU name is "Common KVM processor" or something along those lines, then it means that Windows has detected that it is running in a VM. otherwise, if the CPU tab can detect the name of your CPU and otherwise look normal, then you've succeeded. EDIT: also, KVMs like Unraid and Proxmox can underprovision CPU cores, meaning that you can assign more cores to your VMs than you actually have. this means that you can totally give each of these VMs 16 cores or more, and they'll all run just fine. since most applications and games won't be able to saturate all the cores on a system, you can get away with this quite easily with basically no performance impact. this only becomes an issue if everybody decide to do something stupid, like run prime95 on every VM or something. but even then, the only issue is reduced performance.
  8. people tend to forget that the NVENC feature draws from the same total board power as the rest of the components on a video card. if the card is already maxed out on power draw, enabling NVENC will immediately cause the card to throttle, which leads to lower GPU clock speed, which then leads to reduced performance. likewise, i also use x264 (on my 3950X), and i never get any encoder lag. rendering lag on the other hand, depends on what i'm playing.
  9. and making a video about capture card is the equivalent of selling cocaine? are you listening to yourself? i chose to ignore it when you first brought it up, because it was stupid. but since you decide to double down, i'l bite: if you want to bring up harmful substances into the discussion, then tell me: what is the harmful side effect of a capture card? spoiler: there isn't one. the capture card expands on the capability of your computer with basically zero downsides. the worst that could happen is that you end up not using it much. in which case, if you mistakenly bought one after watching a video that specifically tells you when these things are good idea, then that's on you, not the video. there are cases where this isn't true, which you decide to ignore. WRONG. that's literally the point of having a sponsor spot. the video being sponsored (and clearly marked as such) already signals that there's some amount of bias in place. otherwise, nobody would ever buy sponsor spots. and despite that, and despite your inane "logic", i find the video entertaining and reasonably informative. it covered most of the relevant topics, and actually end up spending very little time actually trying to push a product. in fact, Linus spend a portion of the video pushing one of the sponsor's competitors. how he managed to negotiate that, i have no idea. what you are saying is tantamount to telling Linus that he and his team don't deserve to get paid, or should only be paid according to YOUR terms, which is stupid. i comprehend you just fine. the problem here is that your "logic" makes no sense. the only "logic" you apparently subscribe to, is meaningless nitpicking because you obviously have an axe to grind. grow up and get over it. it's also the one of the only listed use cases brought up by the video. the other one being cameras. nowhere else did it explicitly state that these capture cards are useful for any other use case. the nintendo switch is a modern console, and last i checked, it doesn't have any streaming features. so your statement simply isn't true. and i HAVE actually streamed by streaming my PS4 over wifi/lan to my PC and then streamed the output window using OBS. the experience was miserable. 0/10 do not recommend. i'd definitely recommend a capture card for this scenario.
  10. if showcasing what the product does is considered "favorable light", then sure, i guess he did. except that's not what it means. showcasing what the product does is what Linus did. lol. Linus is sponsored by Corsair and AMD. he spent most of the time talking about Elgato and a little bit of Nvidia. if Linus is a "shill", then he's a pretty terrible one. if i'm Corsair and AMD, i'd consider asking for my money back. where did he lie? what little bit of sponsored content he DID do, he was telling the truth. those Corsair machines with capture cards really are pretty decent for streaming. yes, i even mentioned the fact that some current gen consoles can stream. how well they can stream? well, not great. if you're a serious streamer, you'll need a capture card and a PC that runs it. hence the content of this video. so? did you pop a blood vessel when Linus's channel reviewed a bunch of bicycles? or phones? or vinyl wraps.... for consoles? or heaven forbid... Macs? Linus can review, showcase, do sponsor spots, for whatever he damn well wants. if you aren't stupid, you'd know this. (also lol at the "legacy consoles". he didn't really mention any console in particular, but the one shown on video was a Xbox One of some sort, which last i checked, is current gen. you can't call that "legacy console" when their replacements aren't out on the market.)
  11. ... are you serious? the video RIGHT AT THE BEGINNING clearly stated that it is a sponsored video. you KNOW going into it that it will have product placement. if you fail to realize as a viewer that the video has sponsored content, then that's on you. the rest of your post is pretty meaningless. Linus clearly said, multiple times (as you noticed too) that the capture card is intended to capture footage from external video devices, like consoles and cameras. but, i will highlight one particularly stupid section: consoles? outside of the built-in streaming capabilities found on some consoles, you probably won't be doing any console video capture without a capture card. and what do you know? every bit of game footage shown clearly has a console in the shot. what do you want him to do? mention "console" every 15 seconds? is your attention span that short? i won't go point to point with you, because all of it is nitpicky nonsense, and i don't have time for it. just know that you're more or less wrong on almost all of it. i think you have a just bone to pick to Linus, and you decided to jump onto the first thing that you can sort of attack. feel free to do so, but that'd just make you look sad. the rest of us are just shaking our head at your desperation.
  12. in anticipation to your reply (and replies like it), i just finished watching the video again. the video isn't anything like what you claimed. what i got was a fairly informative video that talked about what capture cards are, their uses, and their limitations, with a little bit of sponsored materials thrown in (none of which really has anything to do with the capture cards themselves). Linus in no way even tried to claim that capture cards are things that you should buy, short of mentioning that on of his sponsors sell them in their products. if viewers somehow get the wrong impression, then that's on them.
  13. really? i'm pretty sure Linus explicitly said that capture cards "are used to record EXTERNAL sources, like a game console, or a camera" within the first 30 seconds of the video. and that's really the purpose of these capture cards: capture video feed from game consoles and cameras. if you just want to stream your computer, you don't need these cards. what I found disappointing, is that Linus failed to mention that for PC recording/streaming, the primary bottleneck for any competent system isn't the encoding part, but the RENDERING part. rendering refers to scene composition, and is run on your GPU core. if your game is stressing your GPU too much, you WILL start dropping frames, and no amount of NVENC/hardware encoders will help you.
  14. sure you do. funny that you mention turns and their resulting pressure drop, when your proposed solution has a whooping SIX turns, each of them near 90 degrees. my build has one. and no, i don't have any blocked off radiators. i may not claim to be a professional engineer, but i'm quite capable of telling whether something is blocked or not. i made a baffle which largely separates the two. it's not perfect, but it should be sufficient in preventing a lot of recirculating airflow you can change my opinion by building this thing and proving it definitively. i do not think that one cannot calculate expected performance beforehand, but i also believe in testing to verify whether something is overlooked or not. build it and prove it, if you can. or not. i don't really care.
  15. the radiators used in that case is 365mm wide. so it's actually about 15-20% of the radiator surface. add in some losses due to tubing runs and whatnot, and 1/10th isn't too far off. you didn't calculate a damn thing - you guestimated. whether any of it translates to results in the real world is anybody's guess at this point. as an example: my personal rig is pretty similar to Linus's, but much jankier: i'm using a modified Rosewill 4U server case with the drive cages drilled out, with two 360mm radiators (largely mounted using packing foam, tape, and prayer). i made an air intake baffle that runs on the bottom of the case to a gap between the two radiators, and the fans on each radiator blows away from each other, drawing the intake air to the front and back. exhaust air are separated from the intake through baffles, so they don't mix. half of the exhaust is set out to the front, so there's less chance of warming up the internal components. layout-wise it's pretty close to Corsair's, where both radiators get fresh air intakes. performance wise? well, its coolant delta T is somewhere around 22C to ambient (which is 20C), with fans running at 1000RPM (6x Noctua NF-A12x25). in custom water cooling, this is generally considered to be pretty terrible. ^ this is why it is important to test your hypotheses, instead of relying purely on fancy drawings and guestimates. we won't know which is actually effective until we test. no. this is about as bad as i imagined it. not to mention, you've made no provision on how to run tubing to this monstrosity, which would reduce airflow even further. i know this is a neat design and all, but i seriously doubt it'll work in the real world, nevermind its obvious space constraints.
×