• 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.

Intel Core Family

What MagicBoy said. It's how many instructions a single core can process in a single clock cycle. It does not magically scale with the number of cores.

Also, if I write a purely single-threaded application, it does not magically make use of additional cores (although underlying OS, drivers, and SDKs might underneath). IPC matters a great deal in these scenarios.
 
ArmitageShanks said:
What MagicBoy said. It's how many instructions a single core can process in a single clock cycle. It does not magically scale with the number of cores.

Also, if I write a purely single-threaded application, it does not magically make use of additional cores (although underlying OS, drivers, and SDKs might underneath). IPC matters a great deal in these scenarios.
Click to expand...

There is no such thing as an application written to use solely one compute thread, other than one or two benchmarking application for academic purposes.

If i was looking at a CPU for Blender you wouldn't recommend an 8600K, actually i think you probably would, and yet the Ryzen 1600 let alone the Ryzen 2600 is the faster CPU, even if you run the 8600K at 5Ghz + the 1600 is still faster.

Now, the 8600K doesn't have SMT but thats besides the point, you wouldn't or rather shouldn't recommend the 8600K to me based on single threaded performance, that's useless to me.

Measuring IPC in reality is not limited to measuring one compute thread only, it assumes, wrongly that a performance of :1 per core scales the same across architectures when the core count is multiplied, the fact that Ryzen has higher performance per clock across multiple cores than Coffeelake relative to it should debunk that.

So if you tell me that the 8700K has 100% performance using one thread compared with the Ryzen 2600 at 97% then i agree Coffelake has 3% higher IPC, and you would like to leave it at that!
Well, IPC is not that restrictive, the 2600 having 106% performance for the same clock rate has the 8700K across multiple compute threads is just as valid, even more valid because nothing actually uses just one core.

Constraining the IPC argument to be subjective only to measure 1 compute thread was and never will be its true intention, it is infact subjective, if you read up on it it will tell you it depends on a myriad of things, including but not limited to, Floating point performance, Integer performance and individual applications, if a CPU has higher performance scaling across multiple compute threads then that is also a valid measurement, the argument that it should be excluded from IPC measurements is a fudge by individuals to exclude anything that doesn't make Intel look the winner.
 
Last edited:
You really would quantify the Ryzen "2" 2600 as having higher IPC than an i7 8700k wouldn't you? @humbug

I'm stunned. But not surprised, as Ryzen gets faster every post you make.
 
Martini1991 said:
You really would quantify the Ryzen "2" 2600 as having higher IPC than an i7 8700k wouldn't you? @humbug

I'm stunned. But not surprised, as Ryzen gets faster every post you make.
Click to expand...

That's because you have a constant habit of assuming a conspiracy of agenda in everything you read.

Blatantly, no, it depends on the use case, as IPC does. how many times have i explained that?

In Cinema 4D it looks like it does, in Blender probably too given it uses the same FP aspect of the CPU. in some other applications i think Coffeelake probably has higher IPC, in games, games bringing you their performance in all sorts of different ways so i think the performance depends on the game.

I do think Cinebech is a good measure of Floating Point performance and a very typical application of it, but no its not the be all and end all.
 
Hate to jump in here,

But humbug has good information and has your or mine interest at heart i think and that is just me.

For example crysis 3 even if it is not the newest direct x version and i have said and if i know more i would sound exactly the same as humbug > https://forums.overclockers.co.uk/posts/31891832

I agree with humbug regarding his comparison of ryzen 2600 vs the 8600k.
 
Martini1991 said:
Luck's got nothing to do with it, I just think I'm an idiot.
Click to expand...

I don't think your an idiot, i agree with you on your statement on about ryzen 2 2600 and 8700k, the 8700k is in a different league.

Either that or i am the idiot, i well and truly think the 8700k has higher ipc compared with the ryzen 2600.

Just speaking my mind.
 
Martini1991 said:
I meant I'm an idiot for replying to Humbug.
Click to expand...

You keep approaching anything i say with the same conspiracy assumptions, so maybe yes.

Anyway, i have a follow-up clip to that original Insurgency CPU cores vid uploading as we speak.
 
humbug said:
You keep approaching anything i say with the same conspiracy assumptions, so maybe yes.

Anyway, i have a follow-up clip to that original Insurgency CPU cores vid uploading as we speak.
Click to expand...

It's not a conspiracy? Ryzen literally does become faster in your posts as the weeks go on. Ryzen "2" you accepted had ~3% IPC improvement only a few weeks ago (After you'd made up a 10% IPC improvement for Ryzen "2") and then suddenly you ignore the past event and make up another figure like you always do. These are factual events. Established facts suddenly disappear, time and time again.

Now we're on to a point where you're making scenarios where you'd consider Ryzen "2" to have faster IPC than Coffeelake. This isn't a conspiracy, you've literally said it today.

You're one of the most AMD biased people on this forum, yet you're the first to shout about "Intel fanboys".
 
Well all i can say is where is the proof... anyways i need to study more i am rusty as i have not been following up on the new processors.

I need to finish what i started,

i said that i would sound exactly the same as humbug well kinda true but i wouldn't shoot at fanboys as for me i am no fanboy i just want the best bang for buck.

I know that i should not engage with putting my nose where it does not belong and that is true,

Anyways i'm out.

Hope this does not turn into a flame war..
 
lltfdaniel said:
Well all i can say is where is the proof... anyways i need to study more i am rusty as i have not been following up on the new processors.

I need to finish what i started,

i said that i would sound exactly the same as humbug well kinda true but i wouldn't shoot at fanboys as for me i am no fanboy i just want the best bang for buck.

I know that i should not engage with putting my nose where it does not belong and that is true,

Anyways i'm out.

Hope this does not turn into a flame war..
Click to expand...

I bought a 1700 and bought a 2700 at launch. Why would I talk down my CPU?

I'm all for buying Ryzen "2". But I wouldn't pretend it's a perfect product.
 
Martini1991 said:
I bought a 1700 and bought a 2700 at launch. Why would I talk down my CPU?

I'm all for buying Ryzen "2". But I wouldn't pretend it's a perfect product.
Click to expand...

Your problem is your preconceptions of me everytime you approach me in a debate, with that you cannot agree with any argument i make, but you also cannot argue with the materiel i use to make my conclusions, so you use anecdotal arguments, "oh i have one and your wrong", well you can make that argument but the on paper facts don't lie.

So yeah maybe just add me to your ignore list.

I'm off to get some lunch.
 
humbug said:
Your problem is your preconceptions of me everytime you approach me in a debate, with that you cannot agree with any argument i make, but you also cannot argue with the materiel i use to make my conclusions, so you use anecdotal arguments, "oh i have one and your wrong", well you can make that argument but the on paper facts don't lie.

So yeah maybe just add me to your ignore list.
Click to expand...

When faced with on paper facts and graphs you only accept it for that day.
The next day or whatnot it's like the event didn't happen.

https://forums.overclockers.co.uk/posts/31842844/

This is when the Ryzen "2" IPC gain over Ryzen 1 was discussed. You'd earlier stated that it was 10% and "backed" it up.

You later accepted it as the ~3% gain I'd originally said it was.

Yet that event didn't happen to you because today you've reverted to your own made up figures with a convoluted method of working it out.

You use flawed evidence, using a graph and then comparing it with a result you have with a maths formula (How you arrived at Ryzen "2" having 10% IPC gain for example). That shouldn't happen in the first place, it's nonsensical to try arguing against it.
 
Last edited:
*removed all but the important part because the forum is caching quotes and readding them after removing them and making a mess of the whole thing.*


Martini1991 said:
When faced with on paper facts and graphs you only accept it for that day.
The next day or whatnot it's like the event didn't happen.

https://forums.overclockers.co.uk/posts/31842844/

This is when the Ryzen "2" IPC gain over Ryzen 1 was discussed. You'd earlier stated that it was 10% and "backed" it up.

You later accepted it as the ~3% gain I'd originally said it was.

Yet that event didn't happen to you because today you've reverted to your own made up figures with a convoluted method of working it out.

You use flawed evidence, using a graph and then comparing it with a result you have with a maths formula (How you arrived at Ryzen "2" having 10% IPC gain for example). That shouldn't happen in the first place, it's nonsensical to try arguing against it.
Click to expand...

Your slide reads like this:

8700K @ 4Ghz = 1325
2600X @ 4Ghz = 1384 (+4.5%)

So in MT Ryzen 2 has 4.5% higher IPC, if you read what i wrote you will see that i guesstimated 6%, i wasn't far off. 1.5% percent off.
 
Last edited:
Please can we stop destroying threads with the same arguments that take on the same track every time and end up personal.

Single core speed is important and a valid metric. Total all thread speed is important and valid. The scaling of more cores/threads is important and valid. Total achievable clockspeed with good air cooling an AIO and a waterloop is valid.

Having an opinion about which is most important or wether a flaw from one side or the other is the defining thing is valid.

I really dislike the security flaws, heat produced and close to requirment to delid to get the best out of an intel. I am still frustrated that AMD are unable to produce a chip that can be pushed to high enough clocks to compete in Single thread tasks even under water and that their chips are so close to max potential at stock.

Does an 8700k beat everything that AMD has for gaming in at least 90% of games when ypu use a top GPU. Definitely, especially if you have the bravery and know how to get the best from it.

Does the improved scaling mean a 2600x beat s an 8700k if you can load all the threads up? I actually dont know if we know that yet and would be really interested to find out if we can come up with some good tests - if it does when is the point that it overtakes? 12 fully loaded threads? 9? That would be cool to know as it would give a serious take on what might happen in the future.

But please can we try and maintain a bit more respect and make the points we want to more plainly wothout bickering definitions? It is getting really old.
 
Beren said:
Please can we stop destroying threads with the same arguments that take on the same track every time and end up personal.

Single core speed is important and a valid metric. Total all thread speed is important and valid. The scaling of more cores/threads is important and valid. Total achievable clockspeed with good air cooling an AIO and a waterloop is valid.

Having an opinion about which is most important or wether a flaw from one side or the other is the defining thing is valid.

I really dislike the security flaws, heat produced and close to requirment to delid to get the best out of an intel. I am still frustrated that AMD are unable to produce a chip that can be pushed to high enough clocks to compete in Single thread tasks even under water and that their chips are so close to max potential at stock.

Does an 8700k beat everything that AMD has for gaming in at least 90% of games when ypu use a top GPU. Definitely, especially if you have the bravery and know how to get the best from it.

Does the improved scaling mean a 2600x beat s an 8700k if you can load all the threads up? I actually dont know if we know that yet and would be really interested to find out if we can come up with some good tests - if it does when is the point that it overtakes? 12 fully loaded threads? 9? That would be cool to know as it would give a serious take on what might happen in the future.

But please can we try and maintain a bit more respect and make the points we want to more plainly wothout bickering definitions? It is getting really old.
Click to expand...

That's nothing to do with anything anyone has said, your reacting to an argument that doesn't exist but on that i completely agree, if you have a GTX 1080TI you should be looking at the 8700K, i have said that everyother post i make and it keeps coming back around to that even if the debate we are having is nothing to do with that.

I said Ryzen 2 has higher IPC in MT in the context that i laid out and provided the materials i based that conclusion on, and no one disagrees with any of it, even Martini provided his own evidence to double, triple down on what i'm saying.....

The 8700K clocks higher, a 5Ghz 8700K is faster than a 4.2Ghz 2600, fact. i don't disagree, i have made that same argument over and over and over again myself.
 
You must log in or register to reply here.
Back
Top Bottom