• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AMD Ryzen 5 5600X Takes the Crown of the Fastest CPU in Passmark Single-Thread Results

Are they gimping intel for the charts? Seems like they fix intel at their base speeds. Something is off here. I would rather wait for proper reviews.


I think the only thing this proves is a lot of intel owners run their chips on crappy motherboards with crappy configurations.....
 
From what I see is they fix all intel cpu's at their base speeds and let 5600x run lose at its normal speeds. Because charts certainly does not state they fix 5600x cpu. If that is true, this is one of the worst misleading and false marketing Inhave witnessed from AMD. Taking people as stupid. But perhaps I'm misreading something and refuse to believe my suspicions for now.
 
It is some 40% faster in 7zip compression. I’m sure there are some other specific loads where the difference is over 20%.
TPU says 15.5 %.
Anyway, I shouldn't have phrased it like that because benchmarks for specific tasksvaries a lot. If looking at overall performance, 21 % is just too much.
In TPU's review the 9900K comes out on top with 3 %.

From what I see is they fix all intel cpu's at their base speeds and let 5600x run lose at its normal speeds. Because charts certainly does not state they fix 5600x cpu.
Unless the benchmark is somehow capable of locking the clock speed on said CPU's I find it highly unlikely, the 9900K benchmark is an average based on 7000 benchmarks made by users.
I think it's just a bad benchmark.
If that is true, this is one of the worst misleading and false marketing Inhave witnessed from AMD.
You think AMD owns passmark? :D
 
At least in Passmark, it's trading blows with the Ryzen 7 3700X even in the full CPU test...
Wow, that's a 75% increase from my 1600X. Even for three generations later, that's very, very impressive (especially considering the 30W TDP drop).
 
There is always a question, how the 5600x will OC. Another question is if the 5600 non-X will launch anyway.
AMD has to offer something around 200 bucks... It's the most commonly bought price range. 5600X is overpriced imho. I'd rather go with 8C i7 10700 (non K) for that money.
 
TPU says 15.5 %.
Anyway, I shouldn't have phrased it like that because benchmarks for specific tasksvaries a lot. If looking at overall performance, 21 % is just too much.
In TPU's review the 9900K comes out on top with 3 %.


Unless the benchmark is somehow capable of locking the clock speed on said CPU's I find it highly unlikely, the 9900K benchmark is an average based on 7000 benchmarks made by users.
I think it's just a bad benchmark.

You think AMD owns passmark? :D

Why do they they desigante intel cpu`s with ` @ xxxx speeds ` and not their single own ? f*** it, I will just wait for the reviews. I do expect it to outperfrom rival product though.
 
Wow, that's a 75% increase from my 1600X. Even for three generations later, that's very, very impressive (especially considering the 30W TDP drop).

You are making my 1600X feel bad now.:oops:

Tbh I'm kinda curious about the cheaper models later, probably next year or so. ~200$ range max.

Those might make me retire the 1600x/B350 after 3 years of using them.
 
AMD has to offer something around 200 bucks... It's the most commonly bought price range. 5600X is overpriced imho. I'd rather go with 8C i7 10700 (non K) for that money.
Judging by their products, and $300 for their 6 core cpu, I expect a 4 core 8 thread cpu at $200 segment. they really upped the prices, and are very confident they will rule the competition. say, ryzen 3 5300x for $ 200.
 
Good job AMD. I honestly never thought I would see the day.

Took you long enough!

Did you ditch the ****ing pins for pads yet?

I always said I would only go back to AMD when they can beat Intel.. in everything.

I wonder if they are stable like a table now..? The last time I rolled with them they were ok for the most part, but not what I'm used to now.
 
TPU says 15.5 %.
Anyway, I shouldn't have phrased it like that because benchmarks for specific tasksvaries a lot. If looking at overall performance, 21 % is just too much.
In TPU's review the 9900K comes out on top with 3 %.
On toms hw the difference is 40%. Maybe they run the processor on stock settings, or the SW version was different? In any case there are cases where the difference is higher than 20%.
 
You are making my 1600X feel bad now.:oops:

Tbh I'm kinda curious about the cheaper models later, probably next year or so. ~200$ range max.

Those might make me retire the 1600x/B350 after 3 years of using them.
Judging by their products, and $300 for their 6 core cpu, I expect a 4 core 8 thread cpu at $200 segment. they really upped the prices, and are very confident they will rule the competition. say, ryzen 3 5300x for $ 200.
There have been rumors of a ~$220 5600 non-X, and frankly it would be really, really weird for AMD not to make such a SKU. $200 for 4c8t even at these performance levels isn't good enough in 2020.
 
There have been rumors of a ~$220 5600 non-X, and frankly it would be really, really weird for AMD not to make such a SKU. $200 for 4c8t even at these performance levels isn't good enough in 2020.

To be honest I would be okay with a decently priced 4/8 say 5300x, not for 200$ tho.
I'm mainly after the single core perf as the games I'm playing rarely multi thread well and I don't do work/CPU heavy related stuff on my PC either.
 
Judging by their products, and $300 for their 6 core cpu, I expect a 4 core 8 thread cpu at $200 segment. they really upped the prices, and are very confident they will rule the competition. say, ryzen 3 5300x for $ 200.
You'd be better off with a 3700x for that price.
New games will like 16 threads, it would be naive to "upgrade" to a 4 core in 2021.
 
To be honest I would be okay with a decently priced 4/8 say 5300x, not for 200$ tho.
I'm mainly after the single core perf as the games I'm playing rarely multi thread well and I don't do work/CPU heavy related stuff on my PC either.
I agree at most a chip like that should be no more than 150. considering, 3300x sold for 125 though you can not get your hands on it for months now...
but intel i3 10320 offering, which is their speediest chip on 4 cores, 8 threads is around 175 now and AMD surely things their range is better and charging more than intel offerings currently, so I would not surprised at $200! not saying, I would buy it, as I wont at that price. but these are the chips that sells and makes the most profit for both companies. not your 400 usd chip.
 
Why do they they desigante intel cpu`s with ` @ xxxx speeds ` and not their single own ?
Their???

This doesn't come from AMD, how many times do I have to tell you? :roll:

On toms hw the difference is 40%. Maybe they run the processor on stock settings, or the SW version was different?
TPU run on stock, and I'd guess Toms does as well.

In any case there are cases where the difference is higher than 20%.
That was my point. There are always extremes, but passmark is supposed (AFAIK) to show overall performance, not extremes. That's a big difference.
Otherwise it would be like saying which graphics card is best based on one gaming benchmark.
 
Last edited:
I agree at most a chip like that should be no more than 150. considering, 3300x sold for 125 though you can not get your hands on it for months now...
but intel i3 10320 offering, which is their speediest chip on 4 cores, 8 threads is around 175 now and AMD surely things their range is better and charging more than intel offerings currently, so I would not surprised at $200! not saying, I would buy it, as I wont at that price. but these are the chips that sells and makes the most profit for both companies. not your 400 usd chip.

3300x was around 160$ in my country before it vanished, 3600 ~200 before the price increase. '~230 nowadays'

It really comes down to the pricing, sure I could always grab a 3600 from the second hand market but thats like my last option preferably. 'This performance increase with the 5000 serie is also hard to ignore'

Not in a hurry tho, saving up for a GPU atm and maybe new CPU around 2021 summer. 'thats when both my mobo+cpu will be 3 years old and thats when I usually upgrade'
 
Wow... You can use it for 10 years!?!o_O

I ran a i7-970 for 10 years wasn't that difficult to do.

So went from Westmere/gulftown and skipped everything after that and went to Zen 2 soon to be Zen 3.

I only upgraded my rig in dec 2019.
 
I ran a i7-970 for 10 years wasn't that difficult to do.

So went from Westmere/gulftown and skipped everything after that and went to Zen 2 soon to be Zen 3.

I only upgraded my rig in dec 2019.

i still have one of my rigs with one running on a Rampage 3 Black edition :D
 
i guess, to upgrade my 3900X i would go 5950X next time ...

Amazing performance for the 5600X, but my 3900X can still get me the full FPS that my monitor outputs (165Hz) :D decisions decision ...
 
There have been rumors of a ~$220 5600 non-X, and frankly it would be really, really weird for AMD not to make such a SKU. $200 for 4c8t even at these performance levels isn't good enough in 2020.
They need to wait for a few months to gather enough low-binned chiplets from the wafers made to have enough chips to start selling that product as it will be sold as hot bread.
 
TPU run on stock, and I'd guess Toms does as well.


That was my point. There are always extremes, but passmark is supposed (AFAIK) to show overall performance, not extremes. That's a big difference.
Otherwise it would be like saying which graphics card is best based on one gaming benchmark.
Toms 5ghz oc result was pretty close to the number 9900k got on TPU. Also the ryzen result is here significantly lower. Anyway, the sw version is different too.

I am with you on that this is a bit different already. Intel has previously dominated these silly microbenches and this indicates that the performance gains are going to be huge.
 
They need to wait for a few months to gather enough low-binned chiplets from the wafers made to have enough chips to start selling that product as it will be sold as hot bread.
Oh, absolutely, they have no reason to push out very cut-down chips early - keeping the highest margin products exclusively available for a few months both allows for building up stocks of lower end, higher volume ones, building up demand by having the most attractive products already out there, and by making some impatient people spend more than they would otherwise. It's a win-win-win scenario for AMD. If it weren't for the price hikes it would also be decent towards users, though with those taken into consideration it's just a clear statement from AMD that "we're not the budget option any more".

Toms 5ghz oc result was pretty close to the number 9900k got on TPU. Also the ryzen result is here significantly lower. Anyway, the sw version is different too.
One possible explanation is motherboard boost settings. What "stock" means here is open to interpretation after all. Intel's specs, the motherboard manufacturers' settings, or something else entirely? Stock on a board with MCE active vs stock on a board where it isn't might be the difference between 5.1GHz all-core and much lower clocks.

Different software versions is also pretty huge though, especially for specific workloads like compression. Might be that one of them has optimizations included that the other lacks.
 
Not really relevant to the news article, but I wonder how long it’ll take for userbenchmark to change its point system (again) after the launch? Will it become just a memory latency test? XD

Rocket Lake is coming and will regain that single-threaded performance crown, despite the fact that in a multi-threaded world the single-thread performance DOES NOT matter.
 
Back
Top