UserBenchmark faces backlash over Ryzen 7 9800X3D review, suggests 13600K and 14600K instead

I have a 14900K along with a 4090 and regardless of resolution 4k or less, there isn't a single game that runs my CPU above 50% at max settings. I bought it to semi future-proof myself and was it a waste of money? Yes, depending on perspective. So does that make buying a top end CPU from AMD a waste of money? Yes.

And, you don't have to deal with AMD software/BIOS issues. No, that isn't a rumor or an old problem that people keep bringing up. It is valid and relevant to this very day. In my professional work 9 times out of 10, if one of my users is having an issue it is an AMD system. Is Intel perfect? Hell no, but I'd trust them for stability LONG before AMD.
 
My rule of thumb for balance at most price points is to spend 2-2.5X on your graphics card compared to your CPU. Unless you're building a cost is no object system, in which case you just buy top of the line all around.

By that rule, if you buy a 9800X3D you should be looking at pairing it with the likes of a 4070 Ti Super or 4080, or a 7900XTX. If your total spend isn't going to be at that level, you'll probably get better gaming performance by stepping down the CPU ladder a rung or two. On the other hand, if you're also going to do non-gaming stuff like video editing you may want to stay with the higher end CPU and sacrifice a few gaming FPS.
 
I have a 14900K along with a 4090 and regardless of resolution 4k or less, there isn't a single game that runs my CPU above 50% at max settings. I bought it to semi future-proof myself and was it a waste of money? Yes, depending on perspective. So does that make buying a top end CPU from AMD a waste of money? Yes.

And, you don't have to deal with AMD software/BIOS issues. No, that isn't a rumor or an old problem that people keep bringing up. It is valid and relevant to this very day. In my professional work 9 times out of 10, if one of my users is having an issue it is an AMD system. Is Intel perfect? Hell no, but I'd trust them for stability LONG before AMD.
No issues with my 3950x, since 2019.

Meanwhile, we've been having to mitigate Intel chipset issues in our fleet of elitebooks at work. The AMD elitebooks work fine.
 
I have a 14900K along with a 4090 and regardless of resolution 4k or less, there isn't a single game that runs my CPU above 50% at max settings. I bought it to semi future-proof myself and was it a waste of money? Yes, depending on perspective. So does that make buying a top end CPU from AMD a waste of money? Yes.

And, you don't have to deal with AMD software/BIOS issues. No, that isn't a rumor or an old problem that people keep bringing up. It is valid and relevant to this very day. In my professional work 9 times out of 10, if one of my users is having an issue it is an AMD system. Is Intel perfect? Hell no, but I'd trust them for stability LONG before AMD.

So you chose documented higher power consumption and chip degradation issues over personal anecdotes about perceived "stability". That's a fine and perfectly valid choice for you, and may even be right! But as someone who spent most of the 2010s on Intel before returning to AMD on their last new build, I can't agree with your assessment. I've been pleasantly surprised at how trouble-free my AM4 system has been for the past 3+ years. Multiple component changes, operating systems, and several thousand hours of gaming. Not one issue.

This is, of course, also anecdotal. Just my two cents. Cheers.
 
And, you don't have to deal with AMD software/BIOS issues. No, that isn't a rumor or an old problem that people keep bringing up. It is valid and relevant to this very day. In my professional work 9 times out of 10, if one of my users is having an issue it is an AMD system. Is Intel perfect? Hell no, but I'd trust them for stability LONG before AMD.
That is an old problem and your timing for this claim is terrible.
Intel 13th and 14th Gen need updates to prevent the CPUs from causing random crashes and protect them from silicon degradation killing the chip.
Arrowlakes launch was a mess with performance all over the place where we'll have to trust Intels word that they'll patch it up.
Meanwhile Intel is cutting costs left and right because they're in financial trouble.

AMD on first gen Ryzen had some issues although I'd say arguably less so than Arrowlake has at the moment. Always a risk with a new architecture.
The only instability issue I can remember was certain USB devices not playing nice with AMDs USB controller (which got resolved with a BIOS update). My gfs PC had that problem with one specific Bluetooth dongle. The laptop, my PC and home server however as worked fine with it.
Other than that AMD has been rock solid for me, my gf and everyone I know with an AMD system. I do however know someone who an unstable 13700k.
Heck, I don't even remember when the home server was updated last, I should look into that. It's been doing its thing without as much a restart since some time in 2023.
 
I have a 14900K along with a 4090 and regardless of resolution 4k or less, there isn't a single game that runs my CPU above 50% at max settings. I bought it to semi future-proof myself and was it a waste of money? Yes, depending on perspective. So does that make buying a top end CPU from AMD a waste of money? Yes.

And, you don't have to deal with AMD software/BIOS issues. No, that isn't a rumor or an old problem that people keep bringing up. It is valid and relevant to this very day. In my professional work 9 times out of 10, if one of my users is having an issue it is an AMD system. Is Intel perfect? Hell no, but I'd trust them for stability LONG before AMD.
it's a shame but you're getting confused pal. That CPU has many e cores/threads, great for industrial multi thread work but not for games where you need 3dcache (smooth, great 1% lows) and you're missing what the cores that ARE being used are actually putting out per core, horsepower if you like (IPC), single core performance numbers. The final nail is obviously how much power intel need to even try to keep up (and therefore the heat that needs extracting from your system)
 
I guess processor reviews on TS are based on using the best hardware out there which currently means the 4090 GPU. This allows the 9800X3D to shine in reviews but it's not a normal situation as most folk can't spend $2K+ on a GPU. Reviews on UB are based on hardware that users own which is likely to include more middle of the range GPUs. In those situations it doesn't make much of a difference what CPU you own. That would be reflected in their reviews.

Perhaps they should allow you to filter the reviews to just owners of high end GPUs and so get a more realistic review of what the CPU is capable of.

 
LOL the stans arguing Intel vs. AMD here are having fun but the point is:

UserBenchmark is full of s**t with their CPU recommendations and has been at least since Ryzen arrived. Changing benchmark weights until Intel comes out on top (the Core i3-10100 was once their top rated CPU, LOL). Histrionic FUD spreading. Hemming and hawing about what makes a CPU better. Whatever they can finagle to make Intel seem the better choice.

Sometimes Intel is the better choice and sometimes it's AMD but not according to UB. It's Intel or nothing for them and their benchmark suite and piles of excuses and whinging will continue to evolve to keep Intel there.
 
Depends on where you come from. Going from a high end AM4 CPU you might not see much of a difference to make it noticeable, but going from an AM3 CPU you could see a significant change.
We aren't coming from AM3 CPUs. Even if we were, the 13600k would still be a better by for 95% of gamers than a 9800X3D.

You're coming from an AM3? Buy a 7600x or a 9700x. Or a 13600k...

The 4090 is a great card, but you shouldn't buy a 4060 because the 4090 is the best card. You shouldn't by an AMD system just because the 9800X3D is the "best gaming CPU".

If you're going to male a "but sometimes!" Argument then we have to talk about "but sometimes!"

You likely aren't a "but sometimes" gamer but you are making the argument. Frankly, I don't feel like putting the work into having this argument. However, if you are then I will.

Give me your best "but sometimes". My best rebuttal to your "but sometimes" is that most people don't have the budget for a 9800X3D if they're still using an AM3 system.
 
Last edited:
We aren't coming from AM3 CPUs. Even if we were, the 13600k would still be a better by for 95% of gamers than a 9800X3D.

You're coming from an AM3? Buy a 7600x or a 9700x. Or a 13600k...

The 4090 is a great card, but you shouldn't buy a 4060 because the 4090 is the best card. You shouldn't by an AMD system just because the 9800X3D is the "best gaming CPU".

If you're going to male a "but sometimes!" Argument then we have to talk about "but sometimes!"

You likely aren't a "but sometimes" gamer but you are making the argument. Frankly, I don't feel like putting the work into having this argument. However, if you are then I will.

Give me your best "but sometimes". My best rebuttal to your "but sometimes" is that most people don't have the budget for a 9800X3D if they're still using an AM3 system.
You said it adds little value to 1440 or 4k. For you, fine. For others, how do you know? You have no argument to stand on other than what you view as acceptable for your needs, not anyone else's.

You can spout numbers all day long from reviews, but in the end what matters is the buyer's opinion. If they find value in it, whatever it may be, then that's good enough for them regardless of what you think.
 
You said it adds little value to 1440 or 4k. For you, fine. For others, how do you know? You have no argument to stand on other than what you view as acceptable for your needs, not anyone else's.

You can spout numbers all day long from reviews, but in the end what matters is the buyer's opinion. If they find value in it, whatever it may be, then that's good enough for them regardless of what you think.
So you're going from objective benchmarks to opinion? Well, that ties up any possibility of an actual discussion that was going to be had on product legitimacy.

Just to really tie things up, I like how you imply my opinion is irrelevant while shifting your argument entirely to consumer opinion.
 
That is an old problem and your timing for this claim is terrible.
Intel 13th and 14th Gen need updates to prevent the CPUs from causing random crashes and protect them from silicon degradation killing the chip.
Arrowlakes launch was a mess with performance all over the place where we'll have to trust Intels word that they'll patch it up.
Meanwhile Intel is cutting costs left and right because they're in financial trouble.

AMD on first gen Ryzen had some issues although I'd say arguably less so than Arrowlake has at the moment. Always a risk with a new architecture.
The only instability issue I can remember was certain USB devices not playing nice with AMDs USB controller (which got resolved with a BIOS update). My gfs PC had that problem with one specific Bluetooth dongle. The laptop, my PC and home server however as worked fine with it.
Other than that AMD has been rock solid for me, my gf and everyone I know with an AMD system. I do however know someone who an unstable 13700k.
Heck, I don't even remember when the home server was updated last, I should look into that. It's been doing its thing without as much a restart since some time in 2023.
I never saw any stability problems with first gen Ryzen, once you paired it with memory it liked. (It was a bit fussy about RAM.) It did have some issues with poor performance if you were doing stuff that fetched memory across from one CCX to another; that was fixed in the second generation, and even before that it was mitigated with a scheduler update that was smarter about which threads it put where. The APUs based on the first generation (2200G and 2400G, and some mobile and embedded chips) never had the issue since they only had one CCX.
 
I know it's basically all but impossible to prove conclusively by us but it seems very convenient for intel that they have at least one high profile site running interference for them, I'm sure they're not going to end up having some sort of inappropriate relationship by getting some sort of sponsorship/advertising deal with intel they'll swear up and down has not affected their unbiased opinion.

Don't think anybody would care to launch another exhaustive investigation into this but in the small chance they do, we might find out interesting things.
Wasn't UB banned from the Intel subreddit?

I'm sure Intel doesn't mind the free advertisement, but I don't think there's any collusion here. The guy who runs UB is just a sourpuss with some kind of grievance against AMD. I don't think he's an Intel shill; he just really hates AMD.
 
I guess processor reviews on TS are based on using the best hardware out there which currently means the 4090 GPU. This allows the 9800X3D to shine in reviews but it's not a normal situation as most folk can't spend $2K+ on a GPU. Reviews on UB are based on hardware that users own which is likely to include more middle of the range GPUs. In those situations it doesn't make much of a difference what CPU you own. That would be reflected in their reviews.

Perhaps they should allow you to filter the reviews to just owners of high end GPUs and so get a more realistic review of what the CPU is capable of.
Techspot/Hardware Unboxed have explained in detail many, many times (including last week) why they (and every other reputable outlet) use (1) the top GPU when benchmarking CPUs and (2) the top CPU when benchmarking GPUs. This is the tried-and-true industry standard.

I recommend watching Hardware Unboxed's latest video on it.

Lastly, even a broken clock is right twice a day. Userbenchmark's data is drivel at best.
 
Last edited:
Wow. Userbenchmark is right on the money and PC Magazine's CPU reviews concur. The X3D CPUS are a niche, compromise. Because the "fastest" gaming performance you will experience is by using an RTX 4090 at 1080P? Who the hell would spend $1500.00-$2400.00 to game at 1080P? Once your gaming is above 1080P you are GPU limited. And the latest 8 core 9800X3D will run about $500.00? If you are lucky?

Userbenchmark is also correct that the Intel 13600 and 14600 provide far more well rounded performance for all applications, not only gaming. Above 1080P, the frame rates for an X800X3D CPU are + or -3% from the rest of the competition.

Also, not all games benefit from vcache anyway. So don't just believe the repeated mantra AMD hopes you will, "X3D best for gaming". Spend your money wisely. Thank you, UserBenchMark for having the courage to tell the truth compared to all the "professional" review sites that use only 1080P with an RTX 4090 GPU for their benchmarks.
First, I'll repeat myself and say that even a broken clock is right twice a day. Either way, UB data is rubbish.

Second, the 9800X3D is currently the best gaming CPU in the world. Period.

Third, none of the established tech outlets are upselling the 9800X3D. They're just stating the obvious, that it is the best gaming CPU. If anything, everyone acknowledges that it does not offer the best productivity performance or the best bang for the buck (I.e., performance per dollar) while gaming. But it does wipe the floor with Intel's top CPU in terms of gaming performance, performance per dollar, and power efficiency.

Finally, nobody here is telling people their CPUs are worthless and need upgrading to the 9800X3D.
 
Last edited:
Well who the heck can afford the Ryzen 7 9800X3D? Is it not overkill for most people? I refuse to buy it because of its price - I would also settle rather for a cheaper CPU.
 
Techspot/Hardware Unboxed have explained in detail many, many times (including last week) why they (and every other reputable outlet) use (1) the top GPU when benchmarking CPUs and (2) the top CPU when benchmarking GPUs. This is the tried-and-true industry standard.

I recommend watching Hardware Unboxed's latest video on it.

Lastly, even a broken clock is right twice a day. Userbenchmark's data is drivel at best.
I understand perfectly why TS uses the hardware it does. I was simply explaining why TS and UB show different results for the same processor.
 
So you're going from objective benchmarks to opinion? Well, that ties up any possibility of an actual discussion that was going to be had on product legitimacy.

Just to really tie things up, I like how you imply my opinion is irrelevant while shifting your argument entirely to consumer opinion.
I didn't go from objective benchmarks. I said it matters on where someone comes from and then you went on to blab about how getting a lesser costing CPU that doesn't match the performance of the 9800x3d was a better route for most gamers.

At 1440p:
The 7700X is slower in gaming by 10%, but falls behind in CPU applications.
The 9700X is slower in gaming by 8% and on par (for the most part) with it in CPU applications.
The 13600k is slower in gaming by 9%, but falls behind in CPU applications.

Everyone's opinion is irrelevant to everyone else - unless they feel it holds merits to their own. Your opinion is just that, your own. If you feel my opinion is moot, that's fine, doesn't hurt my feelings.
 
Back