Thoughts on the benchmarks and reviews?
As expected,
I didn’t do a deep dive on how these tests were done so take it how you will.
Sorta strange Hardware Unboxed (HUB) has different power consumption for loads compared to GamersNexus.
Basically
HUB: 5800x3d > 7800x3d
GN: 7800x3d > 5800x3d
There’s some weirdness with all the reviews, that’s for sure. At least in terms of pure theory it should be an easy win over the 5800X3D as well as the 7700X for the most part, but it just… isn’t. Not always, at least.
At least the 7700X might occasionally see a clock speed advantage, though whether it plays out in practice is questionable with the lower power envelope. There’s no reason it should ever fall behind the 5800X3D, though - higher power envelope, higher clock speeds, and newer architecture shouldn’t result in a regression.
There’s a core parking issue caused by the chipset drivers that I saw covered in one of the reviews (GN I think). Has to do with how the other new X3Ds need to be scheduled affecting the single CCD in the 7800. Wonder how many reviewers checked for it.
Thats ONLY the 7950x3D and the 7900x3D. The 7800x3D is a fantastic chip and has nothing but good reviews. At the moment it’s currently one of the best gaming chips on the market. It is also the most power efficient chip on the market bar none
7800x3d also does NOT EVER park any of the cores the only reason the other chip did it is because only one CCD had the stack die. 7950x3D = 2 8core CCD. 7800X3D= 1 8 core. So the software that handles the core parking is null and void on the 7800X3D. You don’t even need it. This is why the 7800x3D is hands down the chip to get for gamers
With all this being said I do think it’s fair to know as a side note it doesnt mean the 7950x3D is bad. Its just a niche chip. It’s for people who want to have as big a focus on gaming as they do on productivity. So when they use their computer for productivity they have a 16 core great cpu a smigion slower than the 7950x but much more power efficient and when they switch to gaming they have a 8 core v-chace cpu because the 8 non V-cache cores will be parked.
Like I said this is a niche audience
Like I said, it’s a bug. It’s literally covered in the GN video you posted at the 5:07 mark. You are so silly lmao. Debating whether I should just put you on ignore at this point. Please work on your reading comprehension and try watching the stuff you link before you spread more misinformation.
impressive performance with a cost, no denying it’s the CPU to go with if you want to build a top tier performing gaming PC but debatable if it’s worth the money at this current time to people with more modest GPUs and needs that can probably go with something like the AMD 7600x or Intel 13600k.
I don’t pay attention to GN outside of case reviews but both Tom’s and TPU had the 7800X3D more efficient in power use than the 5800X3D (around 15w depending on use)
1 Like
I’m probably going to wait until the dust settles but HUB seems to be the outlier unless I read their graphs wrong.
Seems like the 7800x3d is much more power efficient along with extra bump in performance, but like you said. Gotta pay that bleeding edge tech price tag.
First of all there’s no reason to talk down to someone and give such a rude response. Secondly you seem to have misunderstood what was being said by Steve. The only reason he ran into this issue is because of the fact that he was coming from a 7950x3D with very specific things installed for that chip. Someone who goes out and simply buys the 7800x3D will not run into any of these issues and you’re talking to me about comprehension and misinformation? So yes it is a bug but it’s technically only a bug that testers will face
Steve was quite clear on this and so were a lot of the other reviewers who talked about it unless you’re on a test bench no one is going to be moving from the 7950x3D to 7800x3D. It’s like Leo said in his review disabling game bar fixes this problem when moving from the 7950x3D to the 7800x3D. Disabling game bar completely fixes the issue because that’s the software AMD uses to park cores. People who just go out and get the 7800x3D will not face this issue.
Linus also has a Blog out discussing this. He also clarified that only people on a test bench should face this issue AKA reviewers. I believe I remember Tom’s Hardware mentioning something about it too.
PS> please try and be nicer to people on the forums there’s no reason to give such rude responses to things.
2 Likes
HUB reviews are pretty spot on most times, you can debate their takes sometimes, so maybe there was issue with the mobo or specific test. That said off the top of my head I believe the difference was often 15w between the two. Excellent power efficiency for the 7800X3D but 15w in an extreme test is probably a 10w difference for most real world scenarios and lets face it, if 10-15w is going to make a huge impact on your PC than you may have bigger issues to address.
1 Like
I managed to snag a 7800X3D for MSRP from NewEgg. It arrived yesterday, and I updated my system last night, including moving to Windows 11. After dealing with software frustrations from AMD and Microsoft (like Ryzen Master forcing a Windows repair and BIOS reset), I have everything setup, including an overclock (PBO2 and a -20 offset in the curve optimizer).
Here’s single thread:
vs. 8700k at 5.3GHz:
I also took a screenshot of HW Monitor after typical system usage for a couple of hours. That means Chrome with 2-10 tabs, Discord, YouTube, WoW, and some background utilities to manage fans and GPU overclock. Looking at these numbers, even with the latest version of HW Monitor, I am a bit skeptical of the max boost being accurately recorded. The highest values were reached for just a fraction of a second (and typical max boost was usually 4.8-5.0GHz):
For full specs, the new parts are a Strix B650E-I board, 7800X3D, 32GB 6000MHz CL30 DDR5, and 850W PSU. The parts I kept from the previous build were 2 M.2 drives, a SATA drive, a 3080ti, and a 240mm Fractal Celsius AIO, and I have 5 Noctua fans all in an Ncase M1. The replaced parts were a Strix z370-I board, 8700k, and 16GB 3200MHz CL14 DDR4 with a 600W PSU.
I would have liked to have done more exhaustive testing, but I would have needed to have done a wipe of my old W10 system and installed and updated everything in W11 for better comparison. So what I’ve got are some benchmark data and my perception, which is that the system is snappier overall and that WoW performance is also much smoother, especially in terms of the larger battles (world boss and EBGs) that I frequently find myself in.
1 Like
Griefs I know you’re already well aware of this but I am placing this small explanation here for those reading the thread who are not
It definitely would be much smoother it was the same with the case with the 5800x3d in the case of frame drops. It blew everything else out of the water with higher 1% lows. Which instead of average FPS one the 1% lows is what most people should focus on because that’s what’s going to control how smooth the game feels in the end. Especially when doing things like large group content whether it be PVP or PVE.
World of Warcraft is a very CPU intensive game and it’s one of the games that drink up CPU cache like water in the desert. As I’m sure you’re already aware not all games benefit from more CPU cache but wow was one of those games that most certainly does
Impress that the 7800x3d is able to run roughly the same if not cooler than the 5800x3d.
What’s your fan config for the NCase? All intake with maybe the rear fan as exhaust?
In the T1 I’m running a thermalright x67 as intake with the top fans as exhaust with the T30s. In WoW I averaged mid 70s with the 5800x3d
Ty for the info.
May i ask, how does it perform in crowded areas o fights, like in Valdrakken and Ashran?
1 Like
After looking at parts for a potential build for a few hours I pulled the trigger on everything a minute ago. Definitely overkill for everything I play and hopefully it won’t give me any headaches.
1 Like
1% are as important as avg but not necessarily more important, there are plenty of poor gaming CPUs that give a “smooth” gaming performance because their avg is so low to begin with. Obviously you want to avoid large spikes of FPS and looking at overall performance is important including lows, avg, high as well as how it works with the rest of your system including your video card across all you the games a person plays.
The main thing I am curious about is how it performs in regards to my 5950X in WoW and Factorio
look at Hardware Unboxed review for Factorio.
Am5 (7600x and above)already beat their AM4 counterpart so the 5800x3d and 7800x3d floored them.
It only includes Factorio, I basically need people that play wow to post their experiences since no-one benchmarks WoW