Zando_

Modern GPUs are binned much tighter to begin with and locked down more so this matters less. I don't see any brag about binning on their 7900XTX Nitro+ card either (they usually mention it, did on older Nitro+/Nitro+ LE cards), just the cooler and power delivery being better (power delivery is irrelevant here as the Pulse won't have any issues feeding a 7900XTX to begin with).

I have a Titan XL as well, am ~300lbs, 6' 3". The armrests are crap on mine, the plastic-weld seams on the molded pads have come apart, from what I have heard that is a common issue. The actual seat part itself is built like a tank and I have had 0 issues with it.

The Nitro+ should be a better cooler and the stock clocks are slightly higher. If you don't wanna overclock yourself, and worry about temps a lot, then the $100 may be worth it to ya, otherwise not.

You should push clocks until you're unstable, then drop them by 15-25Mhz and re-test, if you get crashing in some games then drop them a bit further. That'll be the best your card can do, on a 1050 Ti you'll be choked by wattage and voltage limits before you hit a clock limit, the 1050s are locked down tight. No one can give you exact numbers because most AIB cards start at different clocks to begin with, and even on the same model card, different individual GPUs are capable of different clocks (this is called the silicon lottery).

Oh that sounds normal, I'd actually expect it to run lower at idle, probably is when in the OS. That's expected behavior for a CPU with all the power saving settings still enabled (and there's no reason to disable them unless you need to). If you hit it with Cinebench or some other heavy load, you should see it spike up to the proper boost clock. The number Intel gives is for single-core max turbo, by default Cinebench is an all-core workload so you should see a clock ~300Mhz lower than what Intel says the max turbo is. For an inconsistent load like gaming, you'll often see clockspeeds bounce around.

Distilled water is the best coolant possible, we unfortunately have to add stuff to it to keep algae and/or corrosion down, this makes coolant perform slightly worse. Water ain't your problem. If coolant temps hit 100C then your loop would have literally been boiling, and the pressure would have blown the lines loose. What sensor are you getting that data from? How new is the build? It's likely that a bubble got stuck somewhere and messed with the flow. Though I've never had a bubble cause a stoppage to that level. I typically bonk the computer a bit, and if possible lay it on its side, let it run for a bit, then pick it back up, etc. Helps get the air bubbles out.

Hmmm. CPU OC isn't supported on B-boards, RAM OC is though. It looks like it has the VRMs to run a 13900, and they have heatsinks so the VRMs shouldn't be overheating, so that won't be what's throttling the chip. What are you using to check clocks with? HWiNFO64 is the best monitoring tool, CPU-Z if you want to check just the current clocks, most other tools are occasionally inaccurate (especially Task Manager and HWMonitor, which I see lots of people use). What are you running that has it stuck at 2.0GHz, or does it just sit there no matter what load it's under or when it's at idle?

Not overpriced, they're killer value for what they are. Just... what they are isn't a NAS, so I agree that a DIY mATX box is probably better. Something like this: PCPartPicker Part List: https://pcpartpicker.com/list/czkqQP CPU: Intel Core i3-12100 3.3 GHz Quad-Core Processor ($116.49 @ Amazon) Motherboard: ASRock B660M Pro RS Micro ATX LGA1700 Motherboard ($94.99 @ Newegg) Memory: Silicon Power GAMING 32 GB (2 x 16 GB) DDR4-3200 CL16 Memory ($56.97 @ Amazon) Case: Fractal Design Node 804 MicroATX Mid Tower Case ($124.99 @ B&H) Power Supply: Corsair RM750x (2021) 750 W 80+ Gold Certified Fully Modular ATX Power Supply ($119.99 @ Amazon) Total: $513.43 Prices include shipping, taxes, and discounts when available Generated by PCPartPicker 2024-02-14 13:40 EST-0500 Pick drives based off what OS you end up picking (Unraid runs off a USB 2.0 drive, TrueNAS or Proxmox off a standard SSD, or you can run them off a RAID1 SSD setup if you're super worried about a potential failure and allergic to backing up your configuration). Sub out the case depending on space constraints. I'm partial to the Node 804 as it keeps up to 8 3.5" drives on 2 slide-out racks, has 2 more spots for 3.5" or 2.5" drives on the floor of the motherboard side, and 2 2.5" slots in the front panel. There's no 5.25" drive, but if you needed to have a disk drive you can install a laptop-style ODD drive in the front panel as well. It absolutely rules for sheer versatility, but it is a wide cube-style case, so a vertical tower may work better depending on what space you have and where.

For fun? Because they're partnered with Nvidia and want to show off the tech? It's neat tech they wanted to play with and show off. It was never intended to be on for regular gameplay, and that's clearly communicated in the settings where the option exists. It was added after launch, alongside further optimizations to the game (when running without PT of course), so claiming it's a crutch for not wanting to optimize the game is really ignorant. ^^^ I think it's fair to moan a bit about how hard RT can be to run, given it's clearly intended to run with it on. But PT is explicitly a "hey this is neat isn't it" thing, not a core graphics feature.

What CPU/board/RAM is it paired with? What model PSU are you using? The card should pull ~75W from the PCIe slot, putting you around 200-225W... which is still much lower than the 360 or so that a healthy 3080 Ti pulls. This means basically nothing, it just didn't find a stable OC and the limit was voltage, which is always the case on these modern cards as Nvidia does not let you up the voltage much.

It'll be variable, depending on the content, platform, your internet speeds, their server load, etc. There seems to be no list of the various streaming sites, most of the numbers people have are specifically for Netflix. Apple TV is supposedly the highest, but again no numbers. Here's someone who broke down the Netflix numbers, they vary wildly:

Buy the highest res Blu-Rays or DVDs available for the movie/show you want, rip it yourself.

From their page here: "The Folding@home project (FAH) is dedicated to understanding protein folding, the diseases that result from protein misfolding and aggregation, and novel computational ways to develop new drugs in general. Here, we briefly describe our goals, what we are doing, and some highlights so far." F@H is a bunch of simulations to try and figure out why protein folding fucks up, nobody has an estimate of when it will be "done". There isn't a set amount of math to do, as scientists don't know what the problem is, thus why they run folding simulations to begin with. You'll see a lot of "may", "appears to be", "hopefully", etc on their explainers of the various projects for specific diseases. It's a big ole guessing game, really most (all?) science is.

RM's are newer, solid PSUs, should be as good or better than your TX. Fully modular too, so possibly a bit less cable clutter than your TX, depending on what cables you do or don't use.

Yep, unless the fans are going absolutely crazy for no reason, it's fine. My 13" unibody MBP (2012) regularly sat in the 80s, my 15" retina MBP (2015) ran 99C flat under anything harder than downloading a file, fans didn't go silly mode unless it was sustained heavy load. I should say runs, not ran though, as my siblings have the 2012 and a coworker has the 2015 still (it was a work computer, not my personal machine) and both are still kicking. Folks get overly worried about Mac temps, but in my experience they've always ran within Intel spec. Said spec allows the chips to get up to 100C flat which scares people, but it's perfectly fine for a stock Intel chip. The only Mac I've had that wouldn't hop into the 80s or higher immediately was my 2007 Mac Pro, that would sit in the 60s-70s thanks to massive tower coolers. And my M1 MacBook Air now (also a work machine), it doesn't get very hot unless pushed with sustained loads, which I don't really do.

Sliger Cerberus X. Closer to mATX size, but it fits EATX boards and multiple GPUs. Here's an example of one with two 3090s in it: If you want dual x16 GPU slots you'll need HEDT anyways, which usually comes in ATX boards at the smallest (given the massive socket sizes of modern platforms).

5.25" drive and MakeMKV. I dunno how to parse the file system on DVDs either, I just pull the large files as those will be the video ones, the menu stuff is in the megabytes not gigabytes. MakeMKV official free beta key: https://forum.makemkv.com/forum/viewtopic.php?t=1053. If you want to do UHD Blu-Rays as well, then you have to pick drives more carefully: https://forum.makemkv.com/forum/viewtopic.php?f=16&t=19634. I believe regular DVDs are fine with any old DVD drive.

^^^ Admin cmd, diskpart, list disk, sel disk (whatever # is the one you want to wipe), clean (this will nuke the drive, do not run it on something you care about by mistake), exit CMD, yay your drive is back to a single blank and you can format it however you wish.

https://www.truenas.com/community/resources/fake-server-cards.198/ Art of Server has at least one other vid on counterfeits as well. I learned about em when I was looking at what HBA to get for TrueNAS, that's why I picked a single listing card with an authenticity sticker visible on it in the photos, that I could cross-check to confirm it was legit.

Be wary, these are very commonly cloned (same with Intel NICs). I'd avoid listings like that with a bunch, look for one that says in the description that it was pulled from a server, and preferably one with an authenticity sticker you can double-check to confirm it's a real unit. That's what I did for mine, flashed it to IT-Mode myself, it works fine. Clones can be shoddy quality so you can run into issues with them.

1 for the house, 1 for the garage, 1 for the PC desk, 1 for the backpack, seems reasonable.

Plug it in, the software that can take advantage of it will use it. I don't recall if teslas have display outs (my little AMD WX card does), they may be just for acceleration tasks. Which is again, encoding/decoding stuff that GPUs are faster at than CPUs, thus accelerating whatever task they're used for. Here's a success story from Nvidia's site, where folks used the K80s for massive medical visualizations/simulations: https://www.nvidia.com/content/dam/en-zz/Solutions/Data-Center/documents/tu-dresden-success-story.pdf. Yep. If your workload isn't particular to Nvidia's CUDA core setup, then look at AMD cards as they usually have more VRAM for the same price range.

Yep. Both SLI and Crossfire also only work with the same GPU core, so your tesla and a GTX GPU won't work in SLI. As noted, VRAM does not pool in either. Nvidia NVLink on some of their enterprise cards will pool VRAM, but that's about it (and don't confuse NVLink SLI for NVLink, it just uses the NVLink connector for regular SLI, still no VRM pooling). To accelerate graphics. For display out, or encoding/decoding, rendering the viewport in CAD software, etc.

Not quite clear what you're asking. Are you wondering why it pulls more power when under more load? What stress tests are you running?

Looks like a standard USB-C charger, so it should work. I've charged my Airs/Pros with a regular 5-port USB charger using a USB Type A to type C cable, using the USB-C cable from a Caldigit powered dock, using a Jackery Powerbank, and various Apple USB-C chargers. The wattage of the charger will decide how fast the Mac charges, but it should charge. I haven't tried using my Nintendo Switch charger to see if that works, IIRC the USB-C port on the Switch is non-standard, so I don't know if the charger is as well.