Jump to content

Nup

Member
  • Posts

    3,406
  • Joined

  • Last visited

Reputation Activity

  1. Funny
    Nup reacted to Hunrad in The Only Sunbeam 5.25 Bay Cigarette Lighter build of 2023?   
    After a couple of months of waiting for a sketchy Ebay shipment from Latvia, it arrived. The Sunbeam 5.25 bay cig lighter.
    I spent the better part of a year trying to find a legit one. Absolute novelty, works great.
    Anyone else have a PC that still runs one of these?



  2. Like
    Nup reacted to Mihle in Blacks are Grey on Movies in Potplayer, MP-HC/MP-BE on a 10 bit monitor FIX   
    In Potplayer, going to Colorspaces in preferences and selecting RGB 32 in "Output color space" might also be an option that works, but I dont know how that compares to that above, especially when it comes to videos that are not the same as movies usually is (aka for example Full RGB instead of Limited RGB).
  3. Like
    Nup reacted to Mihle in Blacks are Grey on Movies in Potplayer, MP-HC/MP-BE on a 10 bit monitor FIX   
    If you have a 10 bit monitor (or 8 bit+FRC), you might notice that when playing movies in programs like Potplayer or MP-HC or MP-BE, blacks are showing as grey.
    This is at least the case when the display is set to 8 bit in AMD settings, I am not sure about other situations.
     
    The Fix is to install MadVR, a video renderer, and enable it in the video player you choose.

    Before:
    After:
     
    If you dont know how to do it, here is how:
    Dowload MadVR from http://madvr.com Place the folder in a location on your computer where it can stay, you can not delete the folder after setting it up, so you might want to move it out of the dowloads folder. Run the "Install.bat" file in the folder as administrator. (Right click then press run as administrator) Wait for the installation to complete. Enable the Video player to use MadVR:
    Potplayer: MP-BE (Probably the same as MP-HC):
    Restarting the program might be needed, but other than that, Profit!  
    There are options that can be configured in MadVR, but its not needed if your only goal is to fix that the blacks are grey.

    I have no idea why its required when the monitor is set to 8 bit in AMD settings, but it apparently is. If someone else know why, maybe they can explain it down bellow?
     
  4. Like
    Nup reacted to Admiral Shark in Mechanical Keyboard Club!   
    1981 IBM 5281/5282/5285/5286 Data Station Typewriter Keyboard!
     
    * 528X-83 type IBM Model B keyboard
    * 83-key Italian layout
    * IBM beam spring clicky capacitive key-switches
    * SAN double-shot keycaps
    * Made by IBM Italy
    * Shipped with data station terminals of the IBM 5280 Distributed Data System from 1980
    * Some people in the vintage keyboard community call this the "Beamfoot" as it resembles the later and more common IBM 5291/5292 "Bigfoot" Model F buckling spring keyboard but in Model B form
     



  5. Like
    Nup reacted to Volbet in Mechanical Keyboard Club!   
    The groupbuy for the MM Studio Class 80 finally shipped. Decided to populate it with Prevail Bad Seed Tactile switches and a temporary GMK Laser keycap set. I say "temporary," as I wanna try it out with some WoB SA keycaps, but I've yet to find a good set available with a Nordic ISO layout. 
    The only mod I really had to do was the force break mod. Not that I know if it was necessary, but I always do that to keyboards in a aluminium case. 
    Also, this is one hell of a big boy, weighing in at 3,6kg (or almost 8lb) fully build. 
    The GMK set is really nice looks wise, but I don't think I'll ever get the hype surrounding GMK. Mainly because all the sets I've seen, apart from novelty sets, are completely lacking in tertiary characters on the keys. Only primary and secondary characters are printed, so I guess I have to fire blindly when I wanna do a backslash or an at sign.
     
    If you're into retro keyboards, this is definitely a looker, especially if you got the white case for it. 
    Although, the QC on it could certainly have been better. I don't know if this is an issue unique to the solder PCB when run in an ISO layout with a 7U spacebar, but the right Alt-key is essentially swapped with the windows key, meaning that the spacebar covers the right Alt-key, while the cover for the Windows-key actually covers the FN/menu-key. 
    Now, this is incredibly annoying. 
     
    I was able to fix it with VIA and QMK, but it wasn't obvious that the keyboard even supported VIA and/or QMK.  It's not on the MM Studio's website and the keyboard isn't automatically recognized by VIA. You have to go to a third-party seller for the JSON to import it into VIA. 
    That would have been really nice to know before I began wiring the switch on the Windows-key onto the right Alt-key. So while it was fixable (either via software or brute force,) it's small annoyances like this that I had hoped the custom keyboard market had outgrown by this point. 
    Also, and this is more of a VIA and QMK problem, but the split-shift wasn't programmed properly by MM studio. The right side of a split shift, when running as Danish ISO, is supposed to type "<>\," and out of the box the it was set to the US backslash-key, which types " '* " in a Danish ISO layout. I know why this occurs, as that key is used for backslah in other ISO layouts, but it could have been fixed by just assigning the key to the non-US backslah-key in the firmware. 
    Strangely enough, this also isn't the first keyboard I've owned where this is an issue. The NCR keyboard I've posted about earlier suffers from the same issue. 
  6. Like
    Nup reacted to Atk_nite in Mechanical Keyboard Club!   
    I need help opening my Amazon keyboard
    I think I found where they r buuuut theres no screw to unscrew




  7. Like
    Nup reacted to Volbet in Mechanical Keyboard Club!   
    I build this NCR 80 out for a colleague of mine:

     
    I originally intended for it to use Kailh White Owl switches and some white/beige SA profile keycaps, but neither of those things arrived on time. So I guess that will wait for part two.
     
    For now the keyboard uses Drop + Invyr Holy Pandas and Glorious PBT keycaps. The color of the keycaps definitely clash with the retro look of the case, but I do like it in a strange way I can't really describe.  
    It's also weird for the NCR 80 kit to come with LEDs for num lock and scroll lock, considering that this keyboard contains neither key. 
    I probably should also say that the hotswap version of this kit isn't all that easy to build if you use an ISO layout, as you need to move some of the hotswap sockets to other holes by de- and resolder some of them. 
  8. Agree
    Nup reacted to porina in Western Digital Splitting their HDD and Flash Divisions into two Separate Companies   
    For some market context, from first hit on NAND market share - Q2 2023 rounded to nearest %:
    Samsung: 31%
    Kioxia: 20%
    WDC: 15%
    SK Hynix: 18%
    Micron: 13%
    Others: 4%
     
    My memory isn't great, did WD have significant flash offerings before they bought out Sandisk? Splitting HD and flash would be kinda like returning to Sandisk, although that brand only seems to be mainly used for SD cards now.
     
    What are they setting up for by this move?
  9. Agree
    Nup reacted to Taf the Ghost in AMD Radeon RX 7700 XT & 7800 XT Review   
    There's actually a good number of Dubs. Which is good. Since they'll be doing it more, they probably need to think about the dubbing studio's setup a bit. The reflection times in a small booth compared to a large room (normal shooting) means the dubs sound different. In this case, higher reverb and the spatial feel is much closer. Though dubs always have tone differences because the voice changes slightly during the day and between days. Something you don't notice until you put clips back to back.
     
    The dubs definitely made the video flow smoother, which is the point of them. The audio can be fixed a bit with some time to attempt to match it to the normal LMG audio setup, since it'll be more regular.
     
    I think they've used some form of this graphics setup before, right? Either way, I like the new graphics package for the charts. While 1% lows first is a different take than most, it's valid and useful. Glad to also see Production benchmarks hanging around.
     
    I did like the "well, we're waiting" section. AMD's communication on the FSR3 release has been a little lacking. They clearly announced it a tad early or final prep took a little too long. As for working in all games on certain APIs, this isn't the first time AMD as come up with general stuff for their driver package. That's actually been a regular thing. AMD Chill originally only worked in like 30 games until someone on their staff figured out how to get it to work in everything. So I could see most of the stuff they're working working at the driver level like that. We'll just have to see.
  10. Funny
  11. Informative
    Nup reacted to porina in AMD FSR3 demonstrated, along with AFMF   
    There are multiple features at different levels.
     
    Driver level upscaling:
    AMD: RSR
    Nvidia: NIS
    Intel: no news?
     
    AMD are ahead here as NIS feels like a much more basic implementation. Better than nothing but you'd have to be desperate to use it. RSR is comparable to FSR, not FSR 2. I think Intel were first to offer integer (retro) upscaling but that is a niche case.
     
     
    Driver level frame generation:
    AMD: upcoming AFMF, requires RDNA3
    Nvidia: no news
    Intel: no news
     
    Note this is very different from game integrated solutions. The software will not have game engine vector information assist and will have to rely on optical flow. This might be more like "TV quality" frame generation since they work off similar information. PC implementation for gaming will be more latency sensitive.
     
     
    Game integrated upscaling:
    AMD: FSR, FSR2
    Nvidia: NIS, DLSS, DLSS2 - DLSS requires RTX GPUs
    Intel: XeSS (codepaths for Intel and other GPUs)
     
    Recent DLSS is generally considered better than FSR2, although FSR2 remains good enough for most. Even XeSS is starting to gain traction as a multi-vendor option with comparable or better image quality than FSR2.
     
     
    Game integrated frame generation:
    AMD: upcoming FSR 3, minimum RX 5700, recommended 6000 series or newer
    Nvidia: DLSS 3, requires 40 series
    Intel: no news
     
    It is still early days for this tech. Nvidia's implementation has limitations, and AMD's is looking like it will have some too, not necessarily the same ones. On both sides, the FG tech by itself works best for getting games from around 60fps (after upscaling) to higher. It is not going to save 30fps games on console.
     
     
    RT denoising:
    AMD: no news
    Nvidia: upcoming DLSS 3.5 Ray Reconstruction, requires RTX GPUs
    Intel: no news
     
     
    With the impending release of the 7800 XT AMD may finally get some volume going on the current generation sales and we can see how these technologies will help out.
  12. Informative
    Nup reacted to porina in AMD rumoured to have cancelled top die Navi 4C due to complications and also Navi 4C diagram leaked   
    Might be a topic for another thread, how does it compare to EMIB for example? 
     
    I think many have long wondered when we might get GPU core based chiplets as a way to scale higher, although being realistic I think end user cost considerations have overtaken that. Even if we can make effectively bigger GPUs, I don't think most would ever pay for it. For the moment, IMO chiplet based GPU makes most sense at the high end only, nv 80 tier equivalent or higher. Once the packaging matures it could push down the stack further.
     
    The Apple Ultra/Intel Sapphire Rapids way of doing it with fewer bigger dies feels, to me, the more intuitive solution. Current and previous __102 dies have been just over 600mm2. Breaking it down to two 300+ mm2 dies should make it much more comfortable while not significantly increasing packaging complexity. The same die used singly could serve the mid range mainstream too (up to nv 60, maybe 70 tier). For TSMC N5 class I estimate the yields to be 56% at 600 mm2, vs 75% at 300 mm2. This only considers defects, and not binning. Usage of cut down dies would effectively increase the yield so it wont be as bad as it looks.
     
    The solution shown in OP feels to me like breaking it down too far. Individual silicon pieces might be close to 100% yield (current MCD I estimate at >96%), but the packaging complexity is beyond anything we've seen so far, at least for consumer space offerings and even most enterprise.
  13. Informative
    Nup reacted to AlTech in AMD rumoured to have cancelled top die Navi 4C due to complications and also Navi 4C diagram leaked   
    Yes. Navi 44 (replacement for N24 and N33) and N43 (N32 replacement) are expected to be monolithic dies.
     
    There's no N42 or N41 cos of this cancellation.
     
    Speculation is that N44 will be 6nm around 200mm2 die and N43 will be 3nm and be slightly better than 7900XTX perf at significantly reduced power but neither have been leaked or rumoured.
     
    AMD does have big chips but they're for datacenter and AI stuff.
     
    The aim eventually for AMD is to be able to Radeon GPUs the way they make Ryzen CPUs: A truly chiplet based arch instead of the current GPU chiplet approach.
    Because it is cheaper and more efficient to do so. Going bigger is more expensive on monolithic and in the future it will be technically challenging as well.
     
    The reticle limit of a node isn't infinite and it's easier to package small dies into a number of different configurations e.g. using Zen CCDs from Ryzen all the way to Threadripper and Epyc whilst using mostly the same or exactly the same core design except for Zen4c and Zen5c which are different.
  14. Informative
    Nup reacted to porina in AMD rumoured to have cancelled top die Navi 4C due to complications and also Navi 4C diagram leaked   
    I've seen the rumours of no high end AMD GPU next round, but not the why. If this is the "why" it would be an example that not everything AMD works out.
     
    Just on the screenshot alone that's 10 logic pieces and 6 connectivity above substrate, which by itself is already in EPYC territory. It isn't clear how many more repetitions of this there might be in the other dimension. Even if you treat a loaded AID as a subunit, there must be quite some manufacturing (yield) risk to this complexity. The only other example like this I'm aware of might be Ponte Vecchio, at a count of 47 pieces of silicon. But that's HPC, not consumer tier offering.
     
    For lower tier navi 4, I assume they more conventional monolithic route?
     
    For indication, 400mm2 would be comparable to AD103 (4080) at 379, GA104 (3070) at 392, and there isn't a recent AMD GPU in that ball park. NAVI31 (without MCD) is smaller at 304, and NAVI22 (6700) at 335. NAVI21 (6800+) is much bigger at 520. DG2-512 (ARC A700) is also there at 406.
     
    It is also interesting seeing the different approaches the various companies use to get to bigger effective sizes.
     
    Apple M2 Ultra is two dies of ~155 each.
    Intel Sapphire Rapids is 4 dies of 400 each, a total of 1600. It isn't clear to me if EMIB used would count as extra silicon but there are 10 connections.
    Genoa is up to 12x 72 CCDs + 397 IOD, totalling 1261.
     
    AMD certainly is chopping things down to smaller pieces. 
  15. Agree
    Nup reacted to Cyberspirit in Gamers Nexus alleges LMG has insufficient ethics and integrity   
    GN did what should've happened a long time ago.
     
    Everyone was saying that there were a lot of issues with the videos. On this forum and in their comments people always pointed out the issues and voiced their concerns about the videos feeling rushed and unfinished. 
     
    No matter how much people in the community pointed these things out, nothing changed. This reality check had to happen.
     
    If it were to happen "man to man" Linus would've shrugged it off and nothing would've changed. Hell, things might not change either way judging by Linus' response.
     
     
  16. Agree
    Nup reacted to imreloadin in Gamers Nexus alleges LMG has insufficient ethics and integrity   
    The downhill slide for all of this shit started when they started with the clickbait-y titles and thumbnails to "chase the algorithm". *That* was the point they started changing their strategy from relying on the rigor of their content to trying to game the algorithm. I'm not surprised at all that they're making 5+ mistakes a video now as they all struggle to churn out 25+ videos a WEEK. That's an absolutely insane amount of content and something had to give, I'm just sad to see that it was the actual content itself that took the hit.
     
    I wish that Linus would see this for what it is as an eye opener for things that need changed on a cultural level at LMG but he is so deep down the rabbit hole now that all he sees is "moar content = moar algorithm = moar money". It really is sad to see how far LTT has fallen from its roots.
  17. Like
    Nup reacted to dimforest in Gamers Nexus alleges LMG has insufficient ethics and integrity   
    To be fair - it seems more to me that he's defending his own content. You seem to take issue with Hardware Unboxed's response but not the one above it? 
     
    Either way, I don't think this snippet is really helpful to this current discussion regarding ethics and integrity. YouTube comment sections are one of the last places I'd check if I were looking for either.
  18. Like
    Nup reacted to Alan G in Gamers Nexus alleges LMG has insufficient ethics and integrity   
    @LinusTech = I'm sorry but you keep digging a deeper hole rather than owning up to what is really wrong.  If you have a QA/QC problem, heat pumps on the roof are not going to fix that as all.  Your statement about more room for error is just bonkers (to be charitable as I won't use any profanities which is what I really think about this).  You are doing things way to fast and not sitting back and thinking what is the intent of what we are doing with respect to data integrity and presentation.  This is what your whole business depends upon and if you are willing to slough off serious questions about quality, you are really missing what your viewers want and depend upon.  Until you publish a commitment to quality and not just getting bigger and reaching Gary's milestones, LMG's credibility will continue to drop.  You might seriously want to bring some outside consultants in QA/QC to look at all your testing and review procedures as that's the crux of your problem.  If you just want to be a bunch of guys joking around and posting YouTube videos, that's fine with me as I don't watch them but at least you should be honest and state that up front.  The staus quo won't fly!!!!!
  19. Like
    Nup reacted to Wolta in Gamers Nexus alleges LMG has insufficient ethics and integrity   
    The last time I posted on the LTT forums was around the time LTT had just released the first GTX 700 series reviews. I really enjoyed your content in those days.
     
    You’re correct that GN has not followed a journalistic etiquette by allowing the party of topic to respond prior to publication of a critical piece, if you weren’t contacted beforehand.
     
    However, it is impossibly difficult to gloss over the ethos behind this statement; the writer of these words has evidence against them, from a source with historically larger ethos through publication accuracy, that their own professional practices lead open holes for inaccuracies and fallacy in their own product. 
     
    If you found it disappointing that a piece was produced without adhering to etiquette not employed constantly but rather only intermittently by other outlets than The Associated Press, Ritzau, Reuters and the like, imagine the disappointment of us who’ve followed your channel and supported your team monetarily for well over a decade when dishonest practices are revealed - and the response is “Did we make mistakes? Yeah. But we’re humans.”
     
    Repeated offences in reporting accuracy or professionalism costs most of us mortals our jobs. Not just our internet popularity. They are corrected with stringent plans of improvement and outlining how honest mistakes occurred, and how they will be corrected, with future projects safeguarded against these mistakes.
     
    Linus’ reply is not a plan of action or clarification of future intent. It’s an excuse, alongside self-pointed questions, a classic method of deviation away from critical questions so overused it appears as a joke in a certain children’s movie.
     
    5 years ago, when Xiaomi made a MacBook Pro lookalike, LTT and a few other YouTube channels produced content covering it. I bought it off of those reviews’ presentation, implying they were analyses of a product and the analysis showed it was a good product. It turned out to be an insanely flawed laptop, with easily uncovered issues, many of which I since listed in the comments on those videos, if you spent anything more than 24 hours with it. And I got a response to it. 
     
    “Sorry, we just don’t have that amount of time to cover these products so you can’t expect us to do that much.”
     
    I asked back then, and I will ask now 5 years later once again: if covering these products to their exact specification, use case, and the results they deliver in those circumstances is too much work, then why in the world do you brand yourself a reviewer? Why do you talk big game about a “lab” (an insult to anyone having anything to do with actual laboratory science granted the “lab”s flawed results so far…) if you’re not willing to give products and testers the time they need in the first place? Brand yourself as what you are instead - and entertainment influencer - and leave genuine reviews to the organisations that take the time and due diligence to apply scientific standards and proper methodology - and scrutiny, an absolute requirement - to channels that take the extra time to produce this content. 
     
    A such source has called out this unacceptable contrast between branding and actual content delivered by LTT as a content source, and the response is now excuses and self-aimed questions and answers so cliché they’re jokes for kids. 
     
    This is ignoring blatant long term health issues of speeding your employees to a point where they cannot stand behind their own work. This is well documented and meta analyses on the topic are readily available with overwhelming evidence at the US National Library of Medicine’s website. I would know - I wrote my BA on the damned psych. category and designing interventions for it. It’s horrific long-term and killed one of my parents. Shame on you for this alone.
     
    In conclusion, I regret supporting this company entirely and whole heartedly. I’m sorry your day is awful. I hope your company seriously revises itself.
  20. Like
    Nup reacted to rcmaehl in LTT Lemmy Community   
    Not sure if this needs to go into Off-Topic or if General Discussion is fine but it looks like someone setup a LTT community on Lemmy for those of us switching away from reddit
     
    It's available at https://lemmy.ml/c/linustechtips
     
    @7heo this you?
  21. Like
    Nup reacted to Skiiwee29 in Cooler Master ripped off Fan Control   
    Fan Control was actually started by a member of the forum here LONG LONG LONG before Jayztwocents made a video of it. @Rem0o 
     
     
  22. Agree
    Nup reacted to AlTech in Docsis 4 takes a step forward   
    You may want to add a summary of what Docsis is.
  23. Funny
    Nup reacted to colonel_mortis in LTT Forums - 2022 in numbers   
    I'm reluctant to publish other members' stats, but you are indeed #1 for reports filed in 2022, with 531.
  24. Funny
    Nup reacted to MageTank in Noctua Unveils NA-TPG1: A Thermal Paste Guard for AMD AM5   
    Oh it is, lol. I've repasted a ton of times on these AM5 processors and have never had trouble with paste getting in these grooves, and I am known in our lab as the "guy that went to MSI's school of thermal paste application". By that, they mean this:
     
    At worst, there is a small overhang near the grooves that can be easily wiped away, but the SMDs almost never get touched with paste, even with excessive amounts used. This is definitely solving a problem nobody has.
  25. Like
    Nup reacted to mwcarnes22 in This is TWO PCs!   
    Ive got a custom case I designed and had fabricated. Its 4 computers in one case with 40 hard drives.Its done now. These are older images. It actually started its life in an enthoo case.





×