SteveiJobs

I saw these too but I think I'm SOL because this particular card doesn't support driver switching. I also found a very hidden footnote on a product page implying that it required Nvidia vGPU for display output, meaning that it looks like I'm going to be paying for some extra licensing... Oh well, thanks everyone!

Hey everyone, I'm working on an application where I'm passing a GPU (Nvidia RTX 6000 passive) into a VM in ESXi 7u2. After many hours of struggling just getting the VM to boot (hint, there's not much good literature on what the memory mapping value needs to be), I got the VM to boot and installed the drivers and everything is working great! Except... The VM refuses to use the Nvidia GPU and just defaults to the VMWare SVGA 3D driver and won't use the Nvidia card for Blender, games, or anything else that I'm trying to GPU accelerate. I have tried disabling and uninstalling the SVGA 3D driver but no dice (VMWare Tools seems to put it right back). I feel like it might be related to the fact that the card is a server variant with no display outputs and is meant to be used for compute (which it will also be used for on a later project) but all the Nvidia documentation indicates that it can be used for display acceleration for things like Horizon View. Other relevant info: Server is a PowerEdge R7525 with 2x Epyc 7302 CPUs and Guest OS is Server 2019. I tried both the "Datacenter" and "Quadro" drivers from Nvidia. Anybody have any pointers?

CCNA is...fine. I took a lot of prework courses but never the actual exam because it wouldn't have really affected my salary or anything and I have all the info I need for my job. Now a vCenter cert is worth a lot more.

Oh yea you can still move it (sometimes it takes a call into Microsoft to let them know why) but what I was getting at was that the license is tied to the hardware, so you don't even have to enter a key or anything to activate it. Once it's installed, it should automatically recognize the hash in the BIOS and activate.

Maybe. The "Activated with a Digital License" is a function of the key, but also of the mechanism. Certain keys are also activated using KMS or other methods. Microsoft has a lot of "channels" that they sell licenses through (OEM, volume, retail, etc). At the end of the day, the "right" way to license a PC for Windows is to buy a retail copy, but there are lots of other "incorrect, but perfectly legal" ways to buy keys like eBay or resellers that may be getting unused licenses from OEMS, SIs, volume programs, student discount programs, sales, etc. However if you go this way, be warned that it's not supported by Microsoft so if you get a key that's already in use, Microsoft is under no obligation to activate it if you bought it secondhand.

IIRC A digital license is hashed in the motherboard BIOS so if you have to re-install Windows for whatever reason, as long as you don't change your hardware too much, it will just activate no problem. If you have to move motherboards or make a significant hardware change it may not activate though.

I'm just saying that supply and demand are only two variables that affect the price and cost of something. A company like Nvidia manufactures these cards and sets a price based on the margin they make per card and the number they expect to sell. That's MSRP and in a stable market, that's usually the highest it will go for. Something like scalping introduces another variable, competitive purchasing, that artificially inflates demand. If scalpers didn't have bot networks that scooped up cards before retailers could get them, the cards would still go for over MSRP, but nowhere near the 2x-3x markup they're going for now. Scalping bots introduce a competitive disadvantage (just look at all the saltiness in the VAG thread about how it doesn't level the playing field *enough*) and demand rises artificially because of FOMO. Another example would be the Colonial Pipeline hack and a rush on gas in the American Southeast There was a relatively insignificant change in the supply and demand wouldn't have changed if the news never came out, but consumer panic is a real thing. And then there's tariffs, global shipping, market competitors, available alternatives (used cards, APUs), etc. It's a much more complex model than a couple non-mathematical graphs would indicate. At the end of the day, people are frustrated that they cannot get their hands on not just new cards, but even 600/700/900/1000 series cards at reasonable prices. Telling people "suck it up, that's economics" is true, but it's also not helpful. Retailers are trying to find inventive ways to get them in the hands of regular consumers, but the above factors are making that difficult. I don't have an answer (I don't think anybody really does right now), but I empathize with people's frustrations and they are valid.

People who try to use pure Supply and Demand as anything more than a thought exercise are missing the point in my opinion. Supply and demand can't account as easily for how the market actually works (i.e. things like distribution vs direct sales, available competition, marketing, global shipping, etc). For instance, there are different flavors of "demand." Certainly there are gamers out there who want a GPU for their gaming rig. There's also universities building supercomputers for AI/ML, there's miners, there's engineers who need GPU acceleration for Solidworks/AutoCAD, and of course there's bad-actor scalpers who are only buying GPUs for profit. S/D doesn't have a neat explanation for all these different market requirements. Another thing that irks me is on the supply side of the equation because there is definitely a tier of consumers. The supply of GPUs goes to the biggest customer first, which usually means OEMs, supercomputers/universities, and miners simply because they have an enormous amount of buying power. Consumer cards that go to retailers are only a small fraction, and there's always the direct sales on Nvidia's website. TL/DR: yea people need to understand Supply and Demand to make sense of the issue right now, but that's like saying people only need a table saw to build a piece of furniture. It'll get you pretty far but it's not the only tool you need.

One thing I've found working in a tech-adjacent industry is that somebody always wants you to pay for information. Whether that's in real money (training, access to resources like a knowledgebase, software updates) or even in non-monetary resources like time. I've seen a lot of people hold golden information hostage because they think their tribal knowledge gives them job security or a leg up and it's just frustrating because it would help everyone if it wasn't behind some sort of paywall.

Current setup: Windows 10, 2 months old. Old PC: Windows 7 on an HDD, clonezilla'd to an SSD, upgraded to Win10, clonezilla'd to a larger SSD, and decommissioned 2 months ago, 8 1/2 years...

Not sure if this is a local dialect thing (midwest US) but people will use the phrase "I got it offline" to mean they downloaded something from the internet. Another one is when non-tech people cannot understand the difference between downloading and installing software and they just say they "downloaded" it.

This is something I feel like I need to keep clarifying to people. Does Apple collect a lot of telemetry data? Sure. But it's what they do with it that matters. When Apple collects your data on your phone, they generally (to my knowledge at least) keep it mostly in house. Google (and by extension, Android) doesn't because they are an advertising company first, a software company second, and a hardware manufacturer last.

Is there a difference? Sure. But the overarching question is whether it makes any difference to the ear. The numbers you are referencing are resolution and sample rate. For resolution, 16 bit vs 24 bit means that there are 8 more bits in the 24 bit sample than the 16 bit, but at the end of the day, you're just approximating an analog waveform. If the waveform can be reproduced (or captured) in 16 bits worth of data, then there's really no benefit to 24 bit. Same thing goes for sample rate. You can dice the wave up in time into smaller chunks, but is there really an audible difference between 44,000 times per second vs 192,000 times per second? I doubt you could hear the difference. Of course at the end of the day, if the hardware supports it and there isn't a performance penalty, sure, use the higher values. But if you have to lower them for any reason, I doubt it'll make any audible difference and as much as audiophiles like to think they can hear those nuances, they usually can't

I work in industrial automation and it would blow your mind how vehemently people trash Windows 10 (and even Windows 7) just because they learned on XP and can't adapt to a new UI. Even when some of the new features are literally workflow breaking when trying to go back and use XP. For example, simple search in the start menu. Windows 10 (telemetry aside) is actually incredibly stable and I love it as much as a person can love an operating system. (username aside)

Hey everyone! I recently upgrade my TV (had to, the old one died) to a Vizio M series. During this upgrade, I went from 1080P to 4K and I'd like to take full advantage of that but that means upgrading my aging Yamaha receiver. Is there anything to look out for specifically (3.1 setup with a HTPC, Nintendo Switch, RetroPie, GameCube, and turntable connected). I actually really like my current receiver but even aside from the 1080P to 4K thing, I need something with ARC. Thanks!