AMD vs Intel Single threading?

[humbug]

Single core is low and about 1100 below a tuned 9 series. Whether that hurts you is dependent on your workloads.


You also have the ram latency and frequency limitation due to 1:1 ratio to deal with.

I am quite obviously talking about context here, the RAM is junk, the CPU is stock boosting to about 4.1Ghz ST and the Motherboard is an £80 early first gen AM4. I already know some of the guys with better Ryzen 3600 Systems score a lot higher, i don't expect to keep up with systems costing 4 times as much running at way over 5Ghz.

An equivalent Intel priced Intel system is a 9400F and even my very poor example of a Ryzen 3600 is quicker than that.

Probably closer to 17% if both are run on 64-bit or 32-bit. Comparing 32-bit runs against 64-bit runs is a no no unless explicitly wanting to see the difference between them.

People running very old software on Intel's systems to make a point about performance is always suspect to me, especially if they claim as Intel do these days in their clown world that modern real world applications' performance is not what you're supposed to be using to measure performance.

 

[Skyrover]

People running very old software on Intel's systems to make a point about performance is always suspect to me, especially if they claim as Intel do these days in their clown world that modern real world applications' performance is not what you're supposed to be using to measure performance.

It's a good example of raw single core speed which is still hugely relevant in gaming.

 

[humbug]

It's a good example of raw single core speed which is still hugely relevant in gaming.

Which is not that great, its only better, a little, due to the clock speed difference.

The ST IPC on Zen 2 is about is 13% higher than Coffeelake.

Score 535: Intel Core i7 8700K at 5.2Ghz, Chaos666
Score 532: AMD Ryzen R5 3600 at 4.575Ghz, RavenXXX2

Intel needs to run at 5Ghz+ to remain relevant and the thing with that is smaller lithography nodes just don't clock as high, its why they are still with 14nm on desktop, a 5Ghz 14nm Coffeelake CPU its still faster than their next generation architecture is at 4Ghz.

AMD are going a different rout, they are on smaller nodes and with that cannot clock them as high, so they get there with higher per clock performance, and if the rumours are true Zen 3 will have 30% high IPC than Coffeealke, no 5.3Ghz 300 watt 8 core Coffeelake is going to keep up with that.

 

[andy_mk3]

Intel needs to run at 5Ghz+ to remain relevant and the thing with that is smaller lithography nodes just don't clock as high, its why they are still with 14nm on desktop, a 5Ghz 14nm Coffeelake CPU its still faster than their next generation architecture is at 4Ghz.

Yep this is very true and I'm pretty sure the main reason why we have seen so little 10nm from Intel, they simply can't get the clocks out of it to match the very mature 14nm with beefy clocks out the box.

 

[MartinPrince]

I don't remember saying that, i do remember saying that the 3600 and 8700K are probably better for longevity given the thread count difference.

You do write quite a bit so it might be hard to remember exactly what you said but here you go...

No comment on the review slides then? My Zen 2 entry level CPU is faster than your high end Coffeelake, doesn't that sting even just a little?

My 3600 does a better job in Lightroom than his 9700K....!

For this photo editing task yours is 24sec compared to mine at 15secs, that makes mine 37% faster. How's that Humbug pie coming on.

 

[humbug]

You do write quite a bit so it might be hard to remember exactly what you said but here you go...



For this photo editing task yours is 24sec compared to mine at 15secs, that makes mine 37% faster. How's that Humbug pie coming on.

Now you're just flatout trolling... As you did in that thread.

The context of what you just quoted.

The highest ST R20 Ryzen score is 547, this is at 4.7Ghz, the highest Coffeelake score is 580, this is at 5.45Ghz, a score difference of 6% with a clock speed difference of 16%, so yes if you can get past the IPC difference between Zen 2 and Coffeelake then you can beat Zen 2 score, but that IPC difference is very obviously there, if you cannot see that you don't know what IPC is.

On your Lightroom result. I don't use it so i have to concede that to you, i do think its strange that you don't show the filters used for the 9700K.

However, i can use google and the first review i came across has Zen 2 winning.

https://www.pugetsystems.com/labs/a...adripper-2-Intel-9th-Gen-Intel-X-series-1592/

 

[MartinPrince]

Now you're just flatout trolling...

The context of what you just quoted.

Actually, I'm demonstrating real-world photo editing usage (which also mimics most software that does not max out all cores). I have both systems, I use them both every day, I have no foot in either camp and don't spend most of my time trying to big up one platform over the other.

You just seem to find charts and graphs that support your own narrative and then extrapolate to make bold statements that are just ostensibly BS.

You couldn't even remember saying your 3600 was faster than my 9700. You didn't just say 'a 9700' you made a note to mention 'his 9700'. I knew this was BS (if my 3900X isn't then your 3600 is way off), though when I point out your error and remind you of it you start name calling like a puerile child.

I use Photo editing software almost every day (including DXO, Lightroom & Photoshop). You have almost zero experience of using photo editing software and still talk about CS6. I use both a 3900X and a 9700K and the only time that my overclocked/tuned 3900X is faster than my overclocked/tuned 9700K is anything that involves batch work. (where all cores can be loaded)

My 3900X is a brilliant CPU and at the current price OcUK are doing at it is the best overall CPU that can be purchased. Everybody should go out and buy one now! Though when it comes to which is faster in pretty much any task that does not max out all cores then my easily overclocked 9700K is still faster. That is just the simple present truth.

With AMD's next iteration that might all change - I can't wait to see and test for myself, thankfully I won't have to rely on your biased distortions.

 

[Deleted User 456458]

Which is not that great, its only better, a little, due to the clock speed difference.

The ST IPC on Zen 2 is about is 13% higher than Coffeelake.

Score 535: Intel Core i7 8700K at 5.2Ghz, Chaos666
Score 532: AMD Ryzen R5 3600 at 4.575Ghz, RavenXXX2

Intel needs to run at 5Ghz+ to remain relevant and the thing with that is smaller lithography nodes just don't clock as high, its why they are still with 14nm on desktop, a 5Ghz 14nm Coffeelake CPU its still faster than their next generation architecture is at 4Ghz.

AMD are going a different rout, they are on smaller nodes and with that cannot clock them as high, so they get there with higher per clock performance, and if the rumours are true Zen 3 will have 30% high IPC than Coffeealke, no 5.3Ghz 300 watt 8 core Coffeelake is going to keep up with that.

Keep in mind though that those CFL chips aren't just getting that for ST scores but also keeping that all core OC where as the Ryzen chip have no chance of running that frequency at all core. That's why a tuned 3700x/3800x loses to a tuned 9900k.

What AMD needs to do is get their IPC and/or frequency up but also get their all core up. Coupled with a higher IF frequency, Zen 3 should be quite good but they need to improve in multiple areas.

Intel is tapped out on 14nm so anything going forward is just a bit of desperation. They need move off the node to improve.

 

[humbug]

Actually, I'm demonstrating real-world photo editing usage (which also mimics most software that does not max out all cores). I have both systems, I use them both every day, I have no foot in either camp and don't spend most of my time trying to big up one platform over the other.

You just seem to find charts and graphs that support your own narrative and then extrapolate to make bold statements that are just ostensibly BS.

You couldn't even remember saying your 3600 was faster than my 9700. You didn't just say 'a 9700' you made a note to mention 'his 9700'. I knew this was BS (if my 3900X isn't then your 3600 is way off), though when I point out your error and remind you of it you start name calling like a puerile child.

I use Photo editing software almost every day (including DXO, Lightroom & Photoshop). You have almost zero experience of using photo editing software and still talk about CS6. I use both a 3900X and a 9700K and the only time that my overclocked/tuned 3900X is faster than my overclocked/tuned 9700K is anything that involves batch work. (where all cores can be loaded)

My 3900X is a brilliant CPU and at the current price OcUK are doing at it is the best overall CPU that can be purchased. Everybody should go out and buy one now! Though when it comes to which is faster in pretty much any task that does not max out all cores then my easily overclocked 9700K is still faster. That is just the simple present truth.

With AMD's next iteration that might all change - I can't wait to see and test for myself, thankfully I won't have to rely on your biased distortions.

Charts are hard data, they trump anecdotal claims.

 

[humbug]

Keep in mind though that those CFL chips aren't just getting that for ST scores but also keeping that all core OC where as the Ryzen chip have no chance of running that frequency at all core. That's why a tuned 3700x/3800x loses to a tuned 9900k.

What AMD needs to do is get their IPC and/or frequency up but also get their all core up. Coupled with a higher IF frequency, Zen 3 should be quite good but they need to improve in multiple areas.

Intel is tapped out on 14nm so anything going forward is just a bit of desperation. They need move off the node to improve.

You say that as if they are miles apart, a 4.4Ghz Zen 2 is as fast as a 5Ghz Coffeelake.

 

[Deleted User 456458]

You say that as if they are miles apart, a 4.4Ghz Zen 2 is as fast as a 5Ghz Coffeelake.

You can get CFL to 52x 53x all core for 24/7. You're tapped out at 4.4 at best and you gotta be more careful about degrading with voltage due to FIT limitation on a per chip basis. Let's not downplay 200mhz as if Zen 2 all core was 4.6/4.7 we wouldn't hear the end of it

 

[MartinPrince]

Charts are hard data, they trump anecdotal claims.

That hole you're digging for yourself must be over your head by now.

You've said some daft things but this has got to be near the top. I'm not into name calling but before you start banding the 'T' word around you really should take stock.

"Charts are hard data, they trump anecdotal claims". So what about those Principled Technologies Charts released last year? I hear you going "But, but, but....."

This is not just an anecdotal claim, you and I (plus several others) ran exactly the same test. Even taking into account varying OS differences your 37% slower than mine is not even in the same ball park.

 

[humbug]

Even you wouldn't defend Principled Technologies Charts ^^^^^

You can get CFL to 52x 53x all core for 24/7. You're tapped out at 4.4 at best and you gotta be more careful about degrading with voltage due to FIT limitation on a per chip basis. Let's not downplay 200mhz as if Zen 2 all core was 4.6/4.7 we wouldn't hear the end of it

You're beginning to sound like a Bulldozer apologists.

Clocking the pins off your CPU's to keep ahead of the curve is not the way to go, as AMD learned.

Efficient high IPC CPU's is better than high clocked CPU's with very high power consumption needing high end cooling.

 

[MartinPrince]

You say that as if they are miles apart, a 4.4Ghz Zen 2 is as fast as a 5Ghz Coffeelake.

We're talking mainly single threaded usage and anybody with a modicum of AMD knowledge will know it is best not to do an all core overclock on Ryzen and just let the CPU do it's things (maybe enabling PBO).

 

[humbug]

We're talking mainly single threaded usage and anybody with a modicum of AMD knowledge will know it is best not to do an all core overclock on Ryzen and just let the CPU do it's things (maybe enabling PBO).

4.4Ghz all core is not unusual on Zen 2, i set the bar to reasonable levels./

Edit: unless of course they are high core count.

 

[MartinPrince]

4.4Ghz all core is not unusual on Zen 2, i set the bar to reasonable levels./

Edit: unless of course they are high core count.

If you have some kind of learning disability forgive me, (genuinely, as I do voluntary work with people who do) but I don't want to be taking someone to task who doesn't have the ability to stay on point.

I just clearly stated that for single threaded performance (what this topic is about) you really don't want to be using Ryzen all core overclock but yet you reply with some justification of that because seemingly, in your mind it must be that an all core overclock widens the gap between Ryzen and Intel (in Ryzens favour). Not in single threaded tasks, it's the opposite.

My 3900X can boost to 4.650Ghz on 2 cores though I can all core overclock to ~4.350Ghz. If I do an all core overclock my CBr20 Multi-Threaded scores will go up quite a bit but my Single-Threaded score will go down.

So, for the last time, stop mentioning Ryzen All Core overclock in a Single Threaded topic.

 

[humbug]

If you have some kind of learning disability forgive me, (genuinely, as I do voluntary work with people who do) but I don't want to be taking someone to task who doesn't have the ability to stay on point.

I just clearly stated that for single threaded performance (what this topic is about) you really don't want to be using Ryzen all core overclock but yet you reply with some justification of that because seemingly, in your mind it must be that an all core overclock widens the gap between Ryzen and Intel (in Ryzens favour). Not in single threaded tasks, it's the opposite.

My 3900X can boost to 4.650Ghz on 2 cores though I can all core overclock to ~4.350Ghz. If I do an all core overclock my CBr20 Multi-Threaded scores will go up quite a bit but my Single-Threaded score will go down.

So, for the last time, stop mentioning Ryzen All Core overclock in a Single Threaded topic.

Don't start with the passive "Learning Disability" insults.

Singlethreaded or Multithreaded its all the same, the IPC on Zen 2 is 13% higher than Coffeelake, a Coffeelake core needs to run at 5Ghz to keep up with a 4.4Ghz Zen 2 core.

 

[Deleted User 456458]

Don't start with the passive "Learning Disability" insults.

Singlethreaded or Multithreaded its all the same, the IPC on Zen 2 is 13% higher than Coffeelake, a Coffeelake core needs to run at 5Ghz to keep up with a 4.4Ghz Zen 2 core.

What? Single thread favors Ryzen 2 as it can boost higher during those workloads. All core drops core speed so you end up worse while intel equivalent doesn’t.

Ipc is not the same as single core performance. Performance = ipc x frequency.

 

[MartinPrince]

Don't start with the passive "Learning Disability" insults.

You're beginning to sound like a Bulldozer apologists.

So it's OK for you to START throwing insults around at @Robert896r1? Though at the same time you can't appreciate how somebody would seriously question your state of mind after reading through this thread?

I've never seen somebody squirm as much and then jump from one 'sound bite' to the other in an effort to deflect from their own previous stupidly inaccurate statements.

My 3600 does a better job in Lightroom than his 9700K....!

 

[humbug]

You're trolling again ^^^^ i din't Insult @Robert896r1

What? Single thread favors Ryzen 2 as it can boost higher during those workloads. All core drops core speed so you end up worse while intel equivalent doesn’t.

Ipc is not the same as single core performance. Performance = ipc x frequency.

Some people are running them higher all core than they boost single core, it depends.

As for IPC. "Ipc is not the same as single core performance" is a meaningless retort, If you have a given IPC on one core that IPC is the same in each core, if you're measuring core IPC its best to do it on a single core, for Example a 16 thread Zen 2 scales higher in performance than a 16 thread Coffeelake vs what they are on a single core, this is down to better SMT scaling on Zen 2.