Ryzen "2" ?

[gavinh87]

What have you found so far Gavin? I mean you have both a 1700 and a 8700k, why don't you limit your 8700k to the same clockspeed and memory speed/timings and report back to us with some real world testing.

I did 8700k at stock and [email protected]/3466 and 8700k won in every game test I ran. The 1700 is sat on my shelf, cba setting another rig up for benches that have already been done by people with far more credentials than myself.

This isn't about 8700k vs 1700 however, its about ryzen's IPC.
As thestilt has shown ryzen is pretty much at haswell levels when tested over a range of benchmarks.

@humbug There cannot be an absolute without testing every single program out there. They are absolute as far as his own testing goes.

 

[Doobedoo]

I did 8700k at stock and [email protected]/3466 and 8700k won in every game test I ran. The 1700 is sat on my shelf, cba setting another rig up for benches that have already been done by people with far more credentials than myself.

This isn't about 8700k vs 1700 however, its about ryzen's IPC.
As thestilt has shown ryzen is pretty much at haswell levels when tested over a range of benchmarks.

Was the 8700k boosting? Would be interesting though Gavin would be nice to see the results and the exact differences in real world scenarios as opposed to synthetic benchmarks. Nothing wrong with comparing the 1700 to the 8700k as the 2700 will probably be coming out at a similar price to the 8700k if not a little cheaper.

 

[gavinh87]

4core 4thread 3.5ghz testing:
Himeno
Linpack
MCRT
NAMD
Nbody
Nqueen
Stockfish
Winrar compression
X264
X265


gif uploading site

Ryzen 1 vs Coffeelake IPC is as near as makes no difference identical

Do you see how the above tests show that this claim is just wrong? Maybe it has similar IPC in cinebench but it is far from what you claim

 

[gavinh87]

Was the 8700k boosting? Would be interesting though Gavin would be nice to see the results and the exact differences in real world scenarios as opposed to synthetic benchmarks. Nothing wrong with comparing the 1700 to the 8700k as the 2700 will probably be coming out at a similar price to the 8700k if not a little cheaper.

I cannot be 100% sure but it was certainly at stock with MCE off. There will be far better people out there doing the same thing rather than a yam yam with 2 rigs in his spare bedroom lol. The only thing I was interested in was gaming and I think we can all agree the 8700k wins there.

 

[humbug]

I did 8700k at stock and [email protected]/3466 and 8700k won in every game test I ran. The 1700 is sat on my shelf, cba setting another rig up for benches that have already been done by people with far more credentials than myself.

This isn't about 8700k vs 1700 however, its about ryzen's IPC.
As thestilt has shown ryzen is pretty much at haswell levels when tested over a range of benchmarks.

@humbug There cannot be an absolute without testing every single program out there. They are absolute as far as his own testing goes.

If you cannot be absolute because you rightly say you cannot test every program out there then the word "Absolute" has no business being in your article at all, the overriding tone of his article is that his results are absolute indicative of Ryzen IPC, those are the words he used.

If you disagree with the premise that Ryzen IPC is only Haswell then what exactly is your problem my arguments on IPC? don't use some shill thread and argue it meaning, use your won words, your own intelligence.

 

[gavinh87]

ROFL at calling thestilt a shill. Just wow.

This should silence those people to shallow to actually understand what he is trying to show.


He has no problem showing that the 1800x hammers the 7700K in his testing. Remember, he is not testing CPU vs CPU but rather IPC where the conditions are as fair as they can be, I.E no SMT/HT same core count same frequency.

 

[humbug]

So thats two things he didn't use for his "Absolute IPC" chart ^^^^

If you cannot be absolute because you rightly say you cannot test every program out there then the word "Absolute" has no business being in your article at all, the overriding tone of his article is that his results are absolute indicative of Ryzen IPC, those are the words he used.

If you disagree with the premise that Ryzen IPC is only Haswell then what exactly is your problem my arguments on IPC? don't use some shill thread and argue its meaning, use your won words, your own intelligence, like the rest of us.

Come on, drop the clown thread link and make your own case, for once.

 

[Doobedoo]

I cannot be 100% sure but it was certainly at stock with MCE off. There will be far better people out there doing the same thing rather than a yam yam with 2 rigs in his spare bedroom lol. The only thing I was interested in was gaming and I think we can all agree the 8700k wins there.

Yeah we know the 8700k wins in gaming but what we are trying to find out is if it does at the same clock speeds and by how much. Doesn't bother me so much but it would be interesting.

 

[gavinh87]

Yeah we know the 8700k wins in gaming but what we are trying to find out is if it does at the same clock speeds and by how much. Doesn't bother me so much but it would be interesting.

I cannot be certain, and I'm not really willing setting another rig up to find out. I only have a 1070 so extracting the maximum performance of each will be difficult.

 

[CAT-THE-FIFTH]

AFAIK,my impression of Ryzen IPC in non-gaming applications was more between Haswell and Broadwell level. With gaming its more peaky,as it will depend on what kind of engines are used in the said game,and how well it will be optimised for newer designs like Ryzen and Skylake X.

Edit!!

Its why I think that Ryzen "IPC" in gaming will increase over time as more and more devs,will actually bother trying out their new games on Ryzen.

 

[gavinh87]

So thats two things he didn't use for his "Absolute IPC" chart ^^^^



Come on, drop the clown thread link and make your own case, for once.

Clown thread? I don't even know if you are being serious now.
I don't think anyone outside of AMD has put as much work into ryzen than he has. Ryzen RTC checker, yep he made that. Asus' memory overclocking presets, yep thats his too.
I'm pretty sure he holds the overclocking record for the 8370 aswell.


Please do some research before calling someone a shill or their work a clown thread.
Just because you disagree it doesn't make him wrong.

 

[humbug]

Gavin you're still defending him rather than actually addressing my arguments. ./

@Cat, pee take? Haswell and Broadwell are the same ^^^^ 4'th and 5'th gen, the difference between Haswell and Devils Canyon (4770K and 4970K)
--------------

Look, its not that difficult if your willing to put the time in.

Just know some results are with the 8700K at 4.3Ghz and others at 4.7Ghz, depending on the motherboard.

So core for core 1600X vs 8700K if the IPC is the same in games then the 8700K should be between 27% and 39% faster.

1600X: 3.4Ghz vs 8700K @ 4.3Ghz = +27%
Or
1600X 3.4Ghz vs 8700K @ 4.7Ghz = +39%

Now when you take that into account what you will actually see when you look at the performance differences between these two CPU's is that Coffeelake does not have higher IPC in games, it may not even be 'as high'

 

[gavinh87]

No, because you are introducing extra factors here like GPU's etc
Also, you are assuming that things scale linearly, in most cases they do not.
How about we do our own testing here? I'll downclock my 8700k to match your 1600.

 

[CAT-THE-FIFTH]

Gavin you're still defending him rather than actually addressing my arguments. ./

@CAT, pee take? Haswell and Broadwell are the same ^^^^ 4'th and 5'th gen, the difference between Haswell and Devils Canyon (4770K and 4970K)
--------------

Look, its not that difficult if your willing to put the time in.

Just know some results are with the 8700K at 4.3Ghz and others at 4.7Ghz, depending on the motherboard.

So core for core 1600X vs 8700K if the IPC is the same in games then the 8700K should be between 27% and 39% faster.

1600X: 3.4Ghz vs 8700K @ 4.3Ghz = +27%
Or
1600X 3.4Ghz vs 8700K @ 4.7Ghz = +39%

Now when you take that into account what you will actually see when you look at the performance differences between these two CPU's is that Coffeelake does not have higher IPC in games, it may not even be 'as high'

I was looking at comparisons of the Ryzen 5 1600 and 1600X against the Core i7 5820K and Core i7 6800K for non-gaming stuff,and that is with the Intel systems running quad channel memory too.

 

[Doobedoo]

Do it

 

[humbug]

No, because you are introducing extra factors here like GPU's etc
Also, you are assuming that things scale linearly, in most cases they do not.
How about we do our own testing here? I'll downclock my 8700k to match your 1600.

Already on the defensive?

What was that about absolutes? You have been arguing all #### night that Ryzen IPC cannot match coffeelake in games knowing there is no way to know let alone prove it, because you know if you look for the evidence what you will find is for most part the gaming performance difference does not add up to the clock rate difference.

Strawman.

 

[CAT-THE-FIFTH]

Do it

 

[gavinh87]

Already on the defensive?

What was that about absolutes? You have been arguing all #### night that Ryzen IPC cannot match coffeelake in games knowing there is no way to know let alone prove it, because you know if you look for the evidence what you will find is for most part the gaming performance difference does not add up to the clock rate difference.

Strawman.

I'm not being defensive. My only argument is that ryzen does not have coffeelake IPC as you claimed.
We both have 1070's lets clock the CPU's to the same frequency and find out. We can even do 720p to remove any GPU bottlenecks.

 

[humbug]

Doing what exactly?

I'm not being defensive. My only argument is that ryzen does not have coffeelake IPC as you claimed.
We both have 1070's lets clock the CPU's to the same frequency and find out. We can even do 720p to remove any GPU bottlenecks.

That's framed hyperbolic to what i actually said, more Strawman stuff..... what i actually said was this.

Few people do, to be fair most people, including mainstream reviewers still say "Intel has higher IPC" JayZ2Cents whom i respect and like as a reviewer still says it.

To be fair to them in some cases that is true, if you weigh up the single threaded performance in the same Cinenbench you will find Coffeelake is about 8% better, AMD's better than Intel's SMT lets them close that gap to 0 in MT, but again this is still an Intel friendly benchmark, it always was and still is, there are other high load applications where Ryzen is 'significantly' faster than Coffeelake and others where Coffeelake is significantly faster.

If those with the million subscribers actually took the time to look at the IPC question objectively they would find that using a broad brush generalization is wrong, in some instances yes Ryzen has lower IPC, in others Ryzen has higher IPC, mostly they fall somewhere in between, they are different CPU's, one simply cannot broad brush them.

With Mainstream reviewers doing it, what chance does Joe have of knowing whats factually correct?

On the question of gaming. When you actually look at game benchmarks, lots of them, not just one or two outliers what you find is the performance difference is almost always equal to the clock rate differences, sometimes that performance difference is even less than the clock rate difference.

 

[gavinh87]

Ryzen 1 vs Coffeelake IPC is as near as makes no difference identical.

Is what you said. I say lets prove it. I'm sure we have some games with built in benchmarks in common. Or even 3dmark.
720P to remove GPU limits.