Caporegime
Yeah, so with that add 5 to 8% to this result.
And you have Ryzen 2700X real gaming performance in this test.
Which is: 109.2% to 112.3% vs 118.3% with the 8700K.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Ryzen "2" ?
- Thread starter beany_bot
- Start date
Yeah, so with that add 5 to 8% to this result.
And you have Ryzen 2700X real gaming performance in this test.
Which is: 109.2% to 112.3% vs 118.3% with the 8700K.
The difference between MCE and what AMD does is quite simple. AMD guarantees these speeds to work on all CPUs released with no issues while MCE is not guaranteed at all by Intel, it's simply an auto overclock done by the motherboard manufacturers.
While MCE will probably work in most cases, due to it being fairly conservative, it will not in all as evidenced by Gamers Nexus' sample falling over in certain Blender workloads. Also, if you buy and use a basic cooler expecting it to run at stock, MCE could give you a nasty surprise as your CPU will run very hot.
While MCE will probably work in most cases, due to it being fairly conservative, it will not in all as evidenced by Gamers Nexus' sample falling over in certain Blender workloads. Also, if you buy and use a basic cooler expecting it to run at stock, MCE could give you a nasty surprise as your CPU will run very hot.
So you think Canard are saying that running the cpu above the base clock is "cheating" and "auto overclocking"? If you do, i think you may have got the wrong end of the stick.
To me (and others in this thread) he is quite clearly stating that the X470 has auto overclocking on by default, meaning that the all core boost ended up being higher than the rated all core boost at stock (hence the 5-8% advantage).
You seem to be hung up on this whole 3 percent thing, which is only the advantage they measured in games which will always be non linear so drawing conclusions or clocks/ram speeds based on that is a bit silly. Remember, they said it was 15% faster in applications on the A320.
"Compared to its predecessor on the same motherboard with A320 chipset, CanardPC has identified slightly more than three percentage points advantage in games and 15 percent advantage in applications for the Ryzen 7 2700X compared to the Ryzen 7 1800X"
Why didn't they use X370 instead unless you consider AMD's XFR behaviour to cheating even though it's guaranteed by AMD to be stable. It's like trying to disable GPU Boost 3.0 on Nvidia while claiming it's cheating and skewing the results even though it's guaranteed by Nvidia and you lose a lot of performance with it off. This is why I'm confused about what CanardPC have done. We have to wait for more info from reputable sources to find out what's going on.
From a quick google, XFR does work on A320 boards.
Caporegime
Think? i don't need to think CanardPC have done a very good job of explaining it themselves, for a start they make it very clear they did not publish the X470 board results because they thought the boosting was "cheating" its referenced in the links i posted... its also referenced in this simple image, as is the rest of it.
To paraphrase for clarity and feel free to correct me if you read any of this differently.
Q, Juanrga: Using the X470 motherboards instead of the A320 results you published would result in 1 or 2% higher results?
A, CanardPC: in games with Auto overclocking or other cheats enabled you get 5 to 8% higher performance, those cheats were enabled on X470 boards
Q, Juanrga: 5-8% over 1000 series? (in this case 1800X) with auto overcloking?
A, CanardPC: no, over the 2000 series at stock, (in this case the 2700X)
-------------------
Nothing to do with anything other than the fact that they used an A320 board because the X470 was "cheats enabled turbo" its not cheating its just the all core Turbo the Ryzen 2### CPU's were doing by design, they didn't do that on the A320 (probably because its not compatible with Ryzen 2000) and so that's the results they published.
The key here is the 5-8% gaming performance the X470 boards give the 2700X is in CanardPC's mind 'illegal overclocking', "cheating" it isn't, its by CPU design.
Last edited:
Caporegime
ltron kinda gets it.
My GPU is advertised as 1582Mhz Base and 1772Mhz Boost.
It actually runs at between 1898 and 1911 out of the box.
However in this case CanardPC are simply mistaking the 4.35Ghz all core boost as MCE overclocking, it isn't its just the all core boost of the CPU.
If they think the CPU is fine and right running at 4.35Ghz but is 'illegally overclocking' beyond that to gain an extra 8% gaming performance then the Ryzen 2700X is running some really high clocks with that MCE overclocking, clocks the 8700K can only dream of. of course its not doing that.
This is what i tried to explain to you before.
My GPU is advertised as 1582Mhz Base and 1772Mhz Boost.
It actually runs at between 1898 and 1911 out of the box.
However in this case CanardPC are simply mistaking the 4.35Ghz all core boost as MCE overclocking, it isn't its just the all core boost of the CPU.
If they think the CPU is fine and right running at 4.35Ghz but is 'illegally overclocking' beyond that to gain an extra 8% gaming performance then the Ryzen 2700X is running some really high clocks with that MCE overclocking, clocks the 8700K can only dream of. of course its not doing that.
This is what i tried to explain to you before.
Any leak on the European Prices for the 2600 and 2600X?
I ask because I can get now a AB350 ITX Asrock board for 80 euros and a 1600 for 135.
These prices are really good.
And it really makes me wonder if I just just pull the trigger now for my itx build?
Because if 2600 is better but cost 100 euros more than 1600 it won't be worth it so I might as well buy now.
Thoughs or infos?
Cheers!
I ask because I can get now a AB350 ITX Asrock board for 80 euros and a 1600 for 135.
These prices are really good.
And it really makes me wonder if I just just pull the trigger now for my itx build?
Because if 2600 is better but cost 100 euros more than 1600 it won't be worth it so I might as well buy now.
Thoughs or infos?
Cheers!
Soldato
- Joined
- 16 May 2007
- Posts
- 3,221
It will be much more useful after the staged reviews have passed on launch and people generally get their hands on them.
I am interested to know how much better a 2700x performs compared to 1800x and the maximum CPU & RAM speeds achievable. If the 2700x with 16GB of quality RAM gets 4.3Ghz or better and 3600 or better for the RAM that will be a decent enough improvement. The generation that follows is the one I am waiting for.
I am interested to know how much better a 2700x performs compared to 1800x and the maximum CPU & RAM speeds achievable. If the 2700x with 16GB of quality RAM gets 4.3Ghz or better and 3600 or better for the RAM that will be a decent enough improvement. The generation that follows is the one I am waiting for.
Caporegime
At those prices do it, there might be a BIOS update later for 300 series boards to make Ryzen 2000 more compatible, IE the all core Turbo we have just been debating.
Or it might not, but at those prices you're not loosing much if later you fancy upgrading to Ryzen 2000
Soldato
- Joined
- 16 May 2007
- Posts
- 3,221
Think of your budget and what performance you need. Currently they are a bargain for what you get, my 1700 is now available for £100 less that I paid for it a few months after the initial launch.
It just seems to me that CanardPc had (have) the opportunity to show the 2700x in all it's glory but (as expected) they chose not to. Plenty of reasons / excuses, all totally acceptable considering the allegiance of most tech sites to the almighty Intel Pr machine.
The true colours of the 2700x will soon be shown (albeit with every caveat known to mankind) But eventually the end user will know and in that respect maybe then some discerning review sites might be more accommodating.
AMD are being unusually quiet and anticipation is building. For Blue team die hards the chip will always be a failure but in reality it will be another step forward for AMD. Personally I still think there may be some pleasant surprises.
The true colours of the 2700x will soon be shown (albeit with every caveat known to mankind) But eventually the end user will know and in that respect maybe then some discerning review sites might be more accommodating.
AMD are being unusually quiet and anticipation is building. For Blue team die hards the chip will always be a failure but in reality it will be another step forward for AMD. Personally I still think there may be some pleasant surprises.
I can bet you now that 4.35ghz is not the all core boost of the 2700X.
There is no way anyone designs a CPU these days where the all core boost is equal to the single core boost. We had a very credible leak earlier which is being entirely corroborated by this leaked review.
Caporegime
Well i have no reason to argue with that ^^^^ so what? it makes 0 difference to any argument i made.
I'm pretty cynical about reviewers, i have un-subscribed from all the big ones, PCPer, Toms Hardware, Linus Tech Tips..... but there are those out there not yet big enough to have been taken in as a cloaked Intel marketing arm, Hardware Unboxed, Joker Productions, Digital Foundry, JayZ2Cents looks like he's on a conflicted fence tying very hard to be fair to AMD's products while also keeping Intel and nVidia happy, and yes, AdoredTV.
I'm pretty cynical about reviewers, i have un-subscribed from all the big ones, PCPer, Toms Hardware, Linus Tech Tips..... but there are those out there not yet big enough to have been taken in as a cloaked Intel marketing arm, Hardware Unboxed, Joker Productions, Digital Foundry, JayZ2Cents looks like he's on a conflicted fence tying very hard to be fair to AMD's products while also keeping Intel and nVidia happy, and yes, AdoredTV.
Caporegime
How is it? my argument is Ryzen 2### in CanardPC's excluded results is behaving as intended by AMD on the X470 boards, this is contrary to CanardPC's assertions.
Soldato
- Joined
- 16 May 2007
- Posts
- 3,221
Its when you start over clocking it becomes interesting, the 8700k boosts to 4.7Ghz but plenty run theirs on 5.0Ghz all cores overclocked or higher, It is similar for the 8600k. My 1700 is running at its "boost" speed of 3.7 on all cores but would go higher if I wanted to. Older CPU's had more headroom, my previous i5-760 boosted to 2.8Ghz but I ran it overclocked at 3.8Ghz on all cores keeping volts and temp relatively low.
What will be interesting is what all core speed the 2700x can get to without an expensive & extreme cooling solution.
CanardPC state the all core boost of the 2700X is 3.9ghz(+maybe 0.05ghz XFR) and that is what they performed the tests with by using the A320.
What does @humbug believe the all core boost of the 2700X is?
edit:
I mistakenly said 0.5ghz XFR instead of 0.05ghz XFR. Corrected now.
What does @humbug believe the all core boost of the 2700X is?
edit:
I mistakenly said 0.5ghz XFR instead of 0.05ghz XFR. Corrected now.
I don't think even he knows what he thinks it is.
If I pull the trigger on the 1600 now, which of this memory kit is better?
Both are the same price and B-Die Ics.
Patriot Viper DDR4 3733 CL17 19-19-39
GSkill Ripjaws V DDR4 3200 CL14 14-14-34
I know for Ryzen the 3200 CL14 is the sweet spot, but if the 2600 can clock memory higher, is it worth it to go for the 3733 kit even with it's higher latency to future proof and have a memory kit I can later use on a CPU upgrade?
Cheers!
Both are the same price and B-Die Ics.
Patriot Viper DDR4 3733 CL17 19-19-39
GSkill Ripjaws V DDR4 3200 CL14 14-14-34
I know for Ryzen the 3200 CL14 is the sweet spot, but if the 2600 can clock memory higher, is it worth it to go for the 3733 kit even with it's higher latency to future proof and have a memory kit I can later use on a CPU upgrade?
Cheers!