Ryzen vs Coffee Lake

[likeman]

Reading this thread while i was eating popcorn due to to all the entertainment

I see the MCE debated died, amd turbo/xfr/intel turbo is factory, MCE enabled is a overclock its that simple and invalidates websites that with out knowing there cpus was all core boosting to 4.7ghz+extra voltage not sure why there was a debate over it

Also the forum code is screwing around with text 99% of thettime cant replace text now (mobile using SwiftKey ) drivingdme up the wall (note words that are merged) useduto have this problem with lithium websites

 

[juanrga]

I see the MCE debated died, amd turbo/xfr/intel turbo is factory, MCE enabled is a overclock its that simple and invalidates websites that with out knowing there cpus was all core boosting to 4.7ghz+extra voltage not sure why there was a debate over it

No oned here denied that XFR "is factory". And no one denied that MCE "is a overclock". The debate was other.

 

[JustReason]

No oned here denied that XFR "is factory". And no one denied that MCE "is a overclock". The debate was other.

BS. You stated XFR was the same as MCE so that those biased benchmarks you post could be deemed valid.

 

[OrangeKhrush]

BS. You stated XFR was the same as MCE so that those biased benchmarks you post could be deemed valid.

if you can't win, change your debate lol.

 

[drescherjm]

if you can't win

I say neither side one that debate but it was pointless to keep arguing about.

 

[OrangeKhrush]

I say neither side one that debate but it was pointless to keep arguing about.

I think if it was scored on points, the MCE overclock argument equating it to XFR is less points so it would have been a unanimous points decision.

 

[drescherjm]

the MCE overclock argument equating it to XFR is less points

I would not agree with that at all. However I really don't want to continue this. It's not possible that AMD fans will agree with Intel fans on this issue or vise versa.

 

[kirbyrj]

I would not agree with that at all. However I really don't want to continue this. It's not possible that AMD fans will agree with Intel fans on this issue or vise versa.

Then you'd be wrong. This isn't an AMD/Intel debate. This is a debate about what the word "stock" means. Apparently, the Intel fans think all core turbo enabled by the motherboard is a stock configuration. That's what this debate is about.

 

[drescherjm]

Apparently, the Intel fans think all core turbo enabled by the motherboard is a stock configuration.

No, I say they that MCE and XFR are both overclocks. AMD fans want to call XFR as stock. Intel fans want to call it an overclock.

 

[kirbyrj]

No, I say they that MCE and XFR are both overclocks. AMD fans want to call XFR as stock. Intel fans want to call it an overclock.

Might as well lump turbo boost in there also, especially the variable turbo of the CFL. It's the same thing as XFR.

 

[JustReason]

No, I say they that MCE and XFR are both overclocks. AMD fans want to call XFR as stock. Intel fans want to call it an overclock.

not sure what world XFR is considered an OC. It is turbo boost with refined algorithms, but turbo boost none the less and nothing more. MCE is a bypass of that turbo spec StOCK and in no way similar to turbo boost algos used by either AMD nor Intel.

 

[juanrga]

BS. You stated XFR was the same as MCE

I am tired of correcting you:

irrelevant because no one said that XFR is identical to MCE. Repeat. no one.

By eleven time. No one said that "XFR and MCE are the same".

By nth time.... No one said that MCE is XFR.

[Color added to the originals posts]

not sure what world XFR is considered an OC.

In this world. Check #418 and the reviews and slides therein.

 

[juanrga]

Then you'd be wrong. This isn't an AMD/Intel debate. This is a debate about what the word "stock" means. Apparently, the Intel fans think all core turbo enabled by the motherboard is a stock configuration. That's what this debate is about.

Did you even bother to read something of the debate? Because what was claimed is that MCE is an automated overclock...

 

[JustReason]

I am tired of correcting you:







[Color added to the originals posts]



In this world. Check #418 and the reviews and slides therein.

Let me help you here:

You alluded to XFR being akin to MCE and claimed that if MCE is not allowed as a stock setting then XFR could not be used either.

I can make green posts too and for the same reason.

Tell you what state that XFR is just AMDs implementation of Turbo boost and all is well. I mean exactly like this:

XFR is an implementation of Turbo and is not an OC feature like MCE.

No obfuscating or talking out of the sides of your mouth, just state it exactly as I wrote above.

 

[kirbyrj]

Did you even bother to read something of the debate? Because what was claimed is that MCE is an automated overclock...

MCE is like XFR. So it depends of the criteria. I consider that neither XFR nor MCE require a "[OC]" label because both are stock built-in enhancements. People with different criteria must want both XFR and MCE to be labeled as [OC].

What we cannot do is what ADF want. The ADF want to disable MCE on Intel chips, but benchmark RyZen only with the highest interconnect OC possible.

MCE and XFR are stock built-in enhancements.



XFR and MCE aren't exactly the same technology, but both are closely related. From that review that OrangeKrush likes to mention again and again:

XFR is an automatic overclock feature similar to MCE. It is explained in three reviews given above.

RyZen with overclocked RAM would have an "OC" label because overclocking RAM also overclocks the interconnect of the chip. The specs of the chip have been manually changed and the performance varies.

Both XFR and MCE are automatic overclocks. It is not that hard.

I think I started the debate back on page 9 . Plus, did you read your own posts? Of course, you equated XFR and MCE (back on page 9). And then you wonder why we say things like you're being obtuse for no good reason.

I know you won't listen to me, but maybe you'll listen to Anand's take on MCE. In other words definitely a motherboard overclocking feature.

I'll tell you I never said MCE was an automated OC. My point all along is that if it's baked into the CPU (XFR/Turbo Boost) it's not an OC. If it isn't (MCE), it's an OC.

What everyone is missing here is that who even runs their computer with Turbo Boost or XFR enabled anyway when you can run all cores at maximum speed on both AMD and Intel. It's free performance.

 

[juanrga]

Plus, did you read your own posts? Of course, you equated XFR and MCE (back on page 9).

Stating that both MCE and XFR are "automatic overclocks", stating that they are "similar" is NOT equating XFR to MCE. In fact, I described some difference between both technologies.

 

[OrangeKhrush]

Outside of automatic that is about the only similarity, one is on package like turbo and the other is motherboard controlled, MCE also maxes out the most aggressive XMP profile. So if we are going to split hairs automatic is about the only similarity between the two. I am even struggling to find an actual review on XFR, almost like not worth the bother, I don't even think it works unless you are running expensive high end water cooling.

 

[sirmonkey1985]

Outside of automatic that is about the only similarity, one is on package like turbo and the other is motherboard controlled, MCE also maxes out the most aggressive XMP profile. So if we are going to split hairs automatic is about the only similarity between the two. I am even struggling to find an actual review on XFR, almost like not worth the bother, I don't even think it works unless you are running expensive high end water cooling.

XFR is the automatic overclocking which works on my board since my R5 1600 rated at 3.2Ghz stock boots at 3.4Ghz and will turbo boost up to 3.7Ghz. the X series are suppose to have a +100Mhz while the non X have a +50 but for some reason my non X 1600 is +100mhz. there's really not much to talk about though with XFR, either it works or it doesn't.

 

[OrangeKhrush]

XFR is the automatic overclocking which works on my board since my R5 1600 rated at 3.2Ghz stock boots at 3.4Ghz and will turbo boost up to 3.7Ghz. the X series are suppose to have a +100Mhz while the non X have a +50 but for some reason my non X 1600 is +100mhz. there's really not much to talk about though with XFR, either it works or it doesn't.

I haven't seen that on my side, i have seen 3.2 base, 3.4 all core and 3.6 single core which is all defined by AMD themselves. 3.7ghz if assuming that is XFR is then a single thread boost while MCE applies max rated turbo to all cores, which is a massive difference, this is why MT scores very rarely differ on AMD while the MCEgate showed huge difference between the deltas and the MCE results.

Bottom line XFR is completely incapable of affecting results to the point is exists superfluously and the need to even bother testing it is a waste of time, probably why they just OC to 3.9ghz and list that. Juan seems to think because they don't put XFR down on tables it amounts to the same level of shady that MCE gave.

 

[sirmonkey1985]

I haven't seen that on my side, i have seen 3.2 base, 3.4 all core and 3.6 single core which is all defined by AMD themselves. 3.7ghz if assuming that is XFR is then a single thread boost while MCE applies max rated turbo to all cores, which is a massive difference, this is why MT scores very rarely differ on AMD while the MCEgate showed huge difference between the deltas and the MCE results.

Bottom line XFR is completely incapable of affecting results to the point is exists superfluously and the need to even bother testing it is a waste of time, probably why they just OC to 3.9ghz and list that. Juan seems to think because they don't put XFR down on tables it amounts to the same level of shady that MCE gave.

might be a motherboard thing then because my taichi by default sets my 1600 at 3.4Ghz and boosts to 3.7Ghz without having to touch anything even when i was running the stock cooler.

 

[OrangeKhrush]

might be a motherboard thing then because my taichi by default sets my 1600 at 3.4Ghz and boosts to 3.7Ghz without having to touch anything even when i was running the stock cooler.

I don't think it is a motherboard thing as I built my step brother a 1600/Strix combo and Asus/Asrock is essentially the same group. I am curious about that, our one is as packaged CB15 MT runs 3.4ghz, ST runs 3.6ghz like clockwork and without stepping the idle clocks stay at 3.2.

 

[JustReason]

I don't think it is a motherboard thing as I built my step brother a 1600/Strix combo and Asus/Asrock is essentially the same group. I am curious about that, our one is as packaged CB15 MT runs 3.4ghz, ST runs 3.6ghz like clockwork and without stepping the idle clocks stay at 3.2.

My wife's 1600 does the same with a H100i. That 1600 is the most stable result CPU I have ever seen and quite the power miser as well, compared to my [email protected].

 

[bobzdar]

Stating that both MCE and XFR are "automatic overclocks", stating that they are "similar" is NOT equating XFR to MCE. In fact, I described some difference between both technologies.

XFR, being a factory setting, is not overclocking, it's factory stock clocks. There's no such thing as a factory overclock, whatever clocks and boosting comes set up on a factory stock chip are factory stock clocks. Modification to operate above that (ie MCE) is overclocking.

 

[juanrga]

XFR, being a factory setting, is not overclocking, it's factory stock clocks. There's no such thing as a factory overclock, whatever clocks and boosting comes set up on a factory stock chip are factory stock clocks. Modification to operate above that (ie MCE) is overclocking.

#418

 

[bobzdar]

#418

Nowhere does AMD call it an overclock. Nowhere. Anybody who does is wrong. AMD calls it a clock increase, but since it's factory it's stock. Overclock would be anything over XFR or any modification to the XFR profile.

Nothing in your post 418 refutes that.

 

[juanrga]

Nowhere does AMD call it an overclock. Nowhere. Anybody who does is wrong. AMD calls it a clock increase, but since it's factory it's stock. Overclock would be anything over XFR or any modification to the XFR profile.

Nothing in your post 418 refutes that.

Not only AMD description of XFR mentions LN2, but also warns that product warranty does not cover damages caused by overclocking. All this us discussed in post #418.

 

[bobzdar]

Not only AMD description of XFR mentions LN2, but also warns that product warranty does not cover damages caused by overclocking. All this us discussed in post #418.

Which XFR is not - XFR does not void warranty. They mention that you have to have adequeate cooling for XFR to hit peak clocks, but that's with all boost situations, not just XFR. Again, AMD nowhere calls it overclocking and nothing in post 418 refutes that.

 

[thesmokingman]

Must be a slow day in shill land.

 

[JustReason]

Which XFR is not - XFR does not void warranty. They mention that you have to have adequeate cooling for XFR to hit peak clocks, but that's with all boost situations, not just XFR. Again, AMD nowhere calls it overclocking and nothing in post 418 refutes that.

Have you had that feeling yet where you know the person you're talking with cant be that stupid and therefore is just being obtuse for the fun of it? If yes then welcome to the club. I can't get straight answers either.

 

[Pieter3dnow]

I say neither side one that debate but it was pointless to keep arguing about.

There is nothing to argue the fault lies with people not understanding the different technologies one requires a motherboard which has a feature enabled which means the cpu does not function like this on every motherboard and might work if you shell out more money.

I would not agree with that at all. However I really don't want to continue this. It's not possible that AMD fans will agree with Intel fans on this issue or vise versa.

Why the pigeon-holing ?

No, I say they that MCE and XFR are both overclocks. AMD fans want to call XFR as stock. Intel fans want to call it an overclock.

Are they ? if a cheap motherboard does not support MCE where does the performance go?

I am tired of correcting you:







[Color added to the originals posts]



In this world. Check #418 and the reviews and slides therein.

I wish you would be nicer to people by now , you will never hear me say that I grow tired of you , I know you just have a hard time with these complex things (in your case).

Did you even bother to read something of the debate? Because what was claimed is that MCE is an automated overclock...

https://www.anandtech.com/show/6214/multicore-enhancement-the-debate-about-free-mhz

The point of this pipeline post is to ask our readers what they think of MultiCore Enhancement. Do you like it? Does it matter to you? Should it become the standard, or should companies offer different SKUs with and without MCE? If two motherboards from different companies are all equal on price and features but differ by MCE, would you go for MCE? Would you worry about longevity?

There you have it, it is a motherboard feature
Sorry for this post

https://www.techpowerup.com/forums/...-make-it-do-full-4ghz-and-some-4-2ghz.238444/

And this post is from the overclocking and cooling section of the forums so that answers that as well

 

[juanrga]

I wish you would be nicer to people by now , you will never hear me say that I grow tired of you , I know you just have a hard time with these complex things (in your case).

I wish that certain people stop mispresenting my point and stop attributing to me stuff I am not saying. The first time, it could be their misreading or just me not explaining properly my point. The second time, it could be still and error. The third, the fourth, the fiveth... well. You are right on one thing: I just having a hard time to understand why certain people insist on mispresenting my point and continues attributing to me stuff I am not saying.

MCE and XFR are two technologies with similarities and differences. I am mentioning one similarity that they have, but certain people insist on pretending that I "stated XFR was the same as MCE". No I didn't state that.

There you have it, it is a motherboard feature
Sorry for this post

https://www.techpowerup.com/forums/...-make-it-do-full-4ghz-and-some-4-2ghz.238444/

And this post is from the overclocking and cooling section of the forums so that answers that as well

Talking about motherboards, you have in this thread some discussion on how XFR behavior varies with the mobo, even when we consider XFR basically a CPU feature. And about overclocking I found a nice explanation of XFR in Overclockers-UK site. I added their description of XFR for "automatic overclocking" to my resume post #418.

 

[bobzdar]

I wish that certain people stop mispresenting my point and stop attributing to me stuff I am not saying. The first time, it could be their misreading or just me not explaining properly my point. The second time, it could be still and error. The third, the fourth, the fiveth... well. You are right on one thing: I just having a hard time to understand why certain people insist on mispresenting my point and continues attributing to me stuff I am not saying.

MCE and XFR are two technologies with similarities and differences. I am mentioning one similarity that they have, but certain people insist on pretending that I "stated XFR was the same as MCE". No I didn't state that.




Talking about motherboards, you have in this thread some discussion on how XFR behavior varies with the mobo, even when we consider XFR basically a CPU feature. And about overclocking I found a nice explanation of XFR in Overclockers-UK site. I added their description of XFR for "automatic overclocking" to my resume post #418.

Show us anywhere amd calls it overclocking. Anybody else calling it that is wrong. It doesn't void the warranty because its stock, while overclocking does void the warranty.

 

[kirbyrj]

I wish that certain people stop mispresenting my point and stop attributing to me stuff I am not saying. The first time, it could be their misreading or just me not explaining properly my point. The second time, it could be still and error. The third, the fourth, the fiveth... well. You are right on one thing: I just having a hard time to understand why certain people insist on mispresenting my point and continues attributing to me stuff I am not saying.

MCE and XFR are two technologies with similarities and differences. I am mentioning one similarity that they have, but certain people insist on pretending that I "stated XFR was the same as MCE". No I didn't state that.

Stop playing the victim. You have been adamant that XFR was similar enough to MCE to state that MCE benchmarks were valid if XFR benchmarks were valid. When it was pointed out to you that XFR is nothing like MCE for various reasons and is more closely related to Turbo Boost, you went off the deep end.

 

[MaZa]

Reading this thread while i was eating popcorn due to to all the entertainment

I see the MCE debated died, amd turbo/xfr/intel turbo is factory, MCE enabled is a overclock its that simple and invalidates websites that with out knowing there cpus was all core boosting to 4.7ghz+extra voltage not sure why there was a debate over it

Why... WHY did you have to bring it back up...!? Now they are at it again!

 

[juanrga]

Stop playing the victim. You have been adamant that XFR was similar enough to MCE to state that MCE benchmarks were valid if XFR benchmarks were valid. When it was pointed out to you that XFR is nothing like MCE for various reasons and is more closely related to Turbo Boost, you went off the deep end.

Leaving apart that I already stated that the analogous of Intel's Turbo Boost is AMD Precision Boost, and leaving apart that I also discussed the differences between Precision Boost and eXtended Frequency Range, let me recall you that this minor discussion was initiated only as subdiscussion of another more important.

The main discussion was (and continues to be) the double standard of certain people when comparing CoffeLake and RyZen.

I mean that people that takes a review of CoffeeLake with MCE enabled and accuses the journalist of "cheating benchmarks", even when the hardware tested had MCE enabled by default and so correspond to the performance that users will see if purchase that hardware and leave settings unchanged; I mean that same people that link to reviews of Ryzen where the interconnect is overclocked (reducing the CCX-CCX latency), and not only don't accuse them of cheating, but takes those overclocked benches as the performance of the stock chip.

I am simply requiring a fair comparison of RyZen and CoffeLake. Either both chips run on stock configurations or both chips run overclocked (automatically or manually). What I don't accept are those comparisons where the CoffeeLake chip is run fully on stock settings, but the RyZen chip is run with the interconnect overclocked by 20% or more.
And better we forget now those 'reviews' that compare engineering samples of CoffeeLake chips with overclocked RyZen chips, but then label all the chips in graphs as if they were stock.

 

[Stanley Pain]

Stating that both MCE and XFR are "automatic overclocks", stating that they are "similar" is NOT equating XFR to MCE. In fact, I described some difference between both technologies.

Holy shit dude, STFU already. Moving goal posts and all, and now arguing semantics. GTFO.

 

[bobzdar]

Leaving apart that I already stated that the analogous of Intel's Turbo Boost is AMD Precision Boost, and leaving apart that I also discussed the differences between Precision Boost and eXtended Frequency Range, let me recall you that this minor discussion was initiated only as subdiscussion of another more important.

The main discussion was (and continues to be) the double standard of certain people when comparing CoffeLake and RyZen.

I mean that people that takes a review of CoffeeLake with MCE enabled and accuses the journalist of "cheating benchmarks", even when the hardware tested had MCE enabled by default and so correspond to the performance that users will see if purchase that hardware and leave settings unchanged; I mean that same people that link to reviews of Ryzen where the interconnect is overclocked (reducing the CCX-CCX latency), and not only don't accuse them of cheating, but takes those overclocked benches as the performance of the stock chip.

I am simply requiring a fair comparison of RyZen and CoffeLake. Either both chips run on stock configurations or both chips run overclocked (automatically or manually). What I don't accept are those comparisons where the CoffeeLake chip is run fully on stock settings, but the RyZen chip is run with the interconnect overclocked by 20% or more.
And better we forget now those 'reviews' that compare engineering samples of CoffeeLake chips with overclocked RyZen chips, but then label all the chips in graphs as if they were stock.

XFR does not overclock the infinity fabric, so why are you bringing that in?

 

[kirbyrj]

XFR does not overclock the infinity fabric, so why are you bringing that in?

Another red herring. His way of arguing his point is, quite frankly, embarrassing for a fellow Intel owner like myself.

 

[JustReason]

Leaving apart that I already stated that the analogous of Intel's Turbo Boost is AMD Precision Boost, and leaving apart that I also discussed the differences between Precision Boost and eXtended Frequency Range, let me recall you that this minor discussion was initiated only as subdiscussion of another more important.

The main discussion was (and continues to be) the double standard of certain people when comparing CoffeLake and RyZen.

I mean that people that takes a review of CoffeeLake with MCE enabled and accuses the journalist of "cheating benchmarks", even when the hardware tested had MCE enabled by default and so correspond to the performance that users will see if purchase that hardware and leave settings unchanged; I mean that same people that link to reviews of Ryzen where the interconnect is overclocked (reducing the CCX-CCX latency), and not only don't accuse them of cheating, but takes those overclocked benches as the performance of the stock chip.

I am simply requiring a fair comparison of RyZen and CoffeLake. Either both chips run on stock configurations or both chips run overclocked (automatically or manually). What I don't accept are those comparisons where the CoffeeLake chip is run fully on stock settings, but the RyZen chip is run with the interconnect overclocked by 20% or more.
And better we forget now those 'reviews' that compare engineering samples of CoffeeLake chips with overclocked RyZen chips, but then label all the chips in graphs as if they were stock.

Jesus Christ, Lord help this child!!!

MCE is not a function of the CPU therefore a review using it is only illustrating the MoBo utilizing it. XFR is a function of the CPU and varies only by TDP algorithms therefore reviews using it illustrate the CPU. Huge FN difference. You only hate that MCE actually is cheating in a CPU REVIEW and XFR by AMD is quite a bit cooler than Intel turbo boost.

 

[juanrga]

XFR does not overclock the infinity fabric, so why are you bringing that in?

No one said that XFR overclocks the infinity fabric. Infinity fabric is overclocked when overclockiing RAM, as I have explained a dozen of times.

Another red herring. His way of arguing his point is, quite frankly, embarrassing for a fellow Intel owner like myself.

It is indeed embarrassing that your only way of argue consists on continuously misinterpreting my position. Straw man?