Search results

Depends on your current system and how often you are willing to upgrade.

Pretty cool tbh - gaming loads are only in the 50's and 60's, all core only jumps to mid 70's. It may have seen 80C once or twice when I was doing photogrammetry, that's the only situation I've pegged both cpu and gpu at 100% utilization and the only time it runs sustained 100% load on the cpu...

Yep - I waited until prices came down on boards and ram and then jumped in - bonus was they dropped prices of the x3d chips as well. I was surprised at the jump I got from my previous 5800x3d setup tbh. Will wait and see on the next gen, but if we're talking 20% I'll probably have to go for it...

They're good at innovating, not so much on executing...

I'm very lucky that my wife is understanding. I only risk getting beaten when she goes out to the garage....

Updated my sim cockpit with a new frame over the winter and added a 7950x3d. My home built wooden frame wasn't quite holding up to the direct drive wheel and gseat.

Yes, you need to install the chipset drivers, then set to auto in the bios so the drivers can control which ccd things run on.

It's faster than anything but the 7950X in non-gaming workloads, and faster than anything in gaming workloads. It'd be even slower if it had vcache on both dies in non-gaming workloads and no faster in gaming workloads. So it is the best compromise possible with current technology, which is...

This, it's a best of both worlds situation. Basically for clock sensitive apps, they start on the non vcache cores and spill to the lower clocked vcache cores only if needed (in which case the clock drop is much smaller than if they started the other way), while also providing large cache for...

Why have you not pulled the GPU and used the onboard video of the CPU? Have you tried running it with just the cpu, one stick of ram and NO gpu? Then use Q-flash to flash the bios? If that works, then add the gpu back in. If you haven't tried that yet, not sure what you're waiting for....

I don't think the B650's are on the ladder yet, my aorus wasn't but there are two board revisions and I didn't realize it and grabbed the wrong bios and it wouldn't flash, couldn't figure it out until I happened to read the bios version on the box and it didn't match any of the versions listed...

Yes, follow the Q-flash instructions and let it do it's thing. If cpu and memory are coming on then it's not completely dead. Also, try it with no GPU in and the onboard chip. These are very easy so no reason not to do them before sending it in. Make sure you have the correct bios if your...

Yeah, it will depend on the actual card itself (Intel/Realtek etc.) and has to be configured properly. If the Quest connects at a lower speed it will not work well. I have only tried on two and both worked well, but both used the same Intel wifi module iirc.

If your computer has a wifi card, you can turn on hotspot mode and connect the Oculus to it directly and skip the router. This is what I do on my laptop, just make sure it connects on 5ghz and not 2.4ghz (you can toggle in the hotspot options in windows). It will also depend a bit on the wifi...

Um...Techspot and HUB are the same folks (Steve Walton), so the Techspot articles are basically the script for the videos on HUB. And yeah, there are definitely games that benefit from more than 6 cores, but 6 x3d cores are still significantly faster than 8 non-x3d cores...ergo the 7900x3d and...

Yep, it's just another option. 7800x3d if you just game, 7950x3d if you want to game and do workstation type tasks on the same machine and don't care about budget, 7900x3d if you want to do the latter on more of a budget. For $20 more than a 7800x3d it's a really good option tbh. Yeah it...

I mean, any recent game (last 10 years) won't be able to keep the whole game in cache. If the whole thing fits in it will really fly, but that'll only happen in old games that already are pulling hundreds of fps.

2d cards, man. I remember my 486dx2 66 had a Boca 2d card, then a Trident svga in my pentium 120 before moving to 3d accelerators. Before that a Tandy 1000 and ibm ps2 which both had built in video. The ps2 was at least vga graphics. That dx2 66 was the first computer I actually got to do...

I think the real comparison is with the 7950x3d - which is ahead in games and behind in productivity, by about the same amount each. But at half the power. So if you want both high end gaming and productivity prowess, which do you need more of (and are you ok with handling an extra 80-150W of...

I agree with him, if you just want to game, get the 7800x3d - he actually says "you probably should just get the 7800x3d" and I wouldn't even say probably. If you just game, don't get the 7950x3d, it's pointless. Actually in that review CS GO and Valorant are fixed on the 7950x3d, so looks...

s3 Virge +Voodoo1 Riva128 +Voodoo2 Voodoo3 Voodoo4 Radeon 8500 Radeon 9500 (softmod to 9700) 8800GT GTX 480 GTX 1080 RTX 3090FE RTX 4090FE There were a bunch more that were not in my main rig, including a few laptops. The most interesting there was the Acer Helios 5000 AMD with a full desktop...

You obviously didn't get the sarcasm in there. There is no skill required to get a 7950x3d working properly, but apparently you can't seem to manage it...I'll leave the rest, it's been discussed ad nauseam.

Damn, didn't realize every single GPU, motherboard and most usb devices I've ever used are fundamentally flawed and need to be baby-sat with software solutions, too. We should all go back to CPU rendering on only hardware that has native support in windows because the .1% lows were bad in a...

Sure, and 90% are 6+months old. Just like the start of this thread was. And did you also search this? https://www.google.com/search?q=7800x3d+stutter+reddit Again, mostly FUD in here. 7800x3d if you game, 7950X if you work, 7950x3d if you do both and want minimal compromises. If you're...

They do, what are you on about? It's part of the chipset drivers, you have to load them for a 7800x3d system to work optimally, too. What potential performance issues? It outperforms a 7950X in games no matter what. Xbox game bar is installed and enabled by default, so as long as you don't...

I said there's nothing else to troubleshoot. Are chipset drivers installed and game bar enabled. Lots of work there I guess, way more than building 2 systems.

It is free, there is no measurable performance impact, at least wrt to a 7800x3d, which is marginally slower in games due to lower boost. So even if there is overhead it's less than the speed improvement of the higher clocks. As to troubleshooting....I mean what is there to troubleshoot? Are...

There is no babysitting needed, I don't get where this is coming from. For my use, the 7950x3d is superior to either the 7800x3d or 7950X, which is that I game in VR on heavily single thread dependent games and then run 3d photogrammetry and GenAI as a hobby/R&D which need cores. So I get the...

Yeah, exactly, 7950x3d isn't a gaming chip, it's a gaming+productivity chip when you're ok to trade a little bit of productivity performance for maximum gaming performance. That is a pretty niche market tbh, basically gamers that have hobbies or side gigs that can use the extra cores but that...

There are a few glimmers of hope - CP2077 with path tracing really flies in the 7950x3d compared to the 7800x3d, meaning the extra cores help. As we get more path traced and RTX remixed stuff, we may start to see 8 cores bottleneck. A few edge cases now, but by AM6 16 cores might be the new 8....

Heat won't let them have high clocks with v-cache, unfortunately. I think your statement gaming wise was true a year ago when the 7800x3d was released but things have moved on from then and most games no longer ccd hop and stay on the vcache die giving the 7950x3d a small edge almost across the...

Why? 7950x3d is kind of best of both worlds, you get 8 cores with vcache but lower clock speed (but higher than a 7800x3d), 8 cores without vcache but higher clock speeds, so it can handle any workload nearly ideally. If you put vcache on all 16 cores you lose performance in apps that don't...

Is there not an option for setting the curve based on coolant temp if you build your own loop or does it depend on the pump/controller? I know a lot of AIO's have that option, which makes a lot more sense than transient cpu temp given the fan ramp won't immediately cool the cpu as it has to...

I went from a 5800x3d to 7950x3d. I installed latest chipset drivers when doing so, no stutter and improved performance. I mostly game in VR so stutter is usually immediately evident. The 7950x3d cleared up a few frame drops I was getting in iracing while also improving 3d photogrammetry...

It moves loads around from core to core to better distribute heat and allow higher overall boost, that's why you're seeing it loaded even on less than all core loads, but it should run least often on that core.

What's the core ranking (you can see this in hwinfo64)? It's probably already the lowest ranked core so would be used least by default.

You can measure performance on a large language model in initial response time and tokens per second. I did a bunch of testing on hardware I have and will test the 7950x3d when it gets here this weekend: Asus G14 (16gb lpddr4 3200mhz) 4900HS IGP (Vega 8CU) 1.5 tok/s 4900HS CPU (Zen 2 8 cores)...

I'm using LM studio to host local models and TaskWeaver and/or OpenInterpreter for interactions - basically having them write and test code for different tasks. I work in the AI/ML space and we can't use remotely hosted models due to data sharing concerns so being able to test them locally...

I was originally planning to skip this generation - but I wanted to go to 128gb of ram for AI inference and didn't want to invest more in AM4. I already have some use cases that benefit from 16 cores, though not enough to make me upgrade originally to a 7950x3d, and still game quite a bit, so...

Update - GPU inference is slower than CPU (5 tok/s vs. 14 tok/s). I tested against my 5800x3d and it only got around 9 tok/s, so the ddr5 seems to be making a huge difference on the cpu side for AI inference. This was using a larger model (17GB in ram). For reference, my 4090 gets 80 tok/s...