• 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.

AMD vs Intel Single threading?

MartinPrince said:
This same CPU scored 528 in a different motherboard which would only boost to 4575Mhz but in this one that normally boosts to 4625Mhz it scored sightly less, which was not what I expected. On the plus side this motherboard scores 200 points higher in multicore at stock, which is higher than most I've seen. As this is mainly used for multicore work I'm happy with that.

49207338623_9b23cbe17d_c.jpg
Click to expand...

So when you said your tuned 3900X for this particular test of DXO was actually not, right? The 9700 at 5.2 was actually tuned but no mention that it was.
 
amigafan2003 said:
I'm interested to see how mine compares against a 9700k - plse send me the details - already have DXO installed.
Click to expand...
Here is the link to the RAW file
https://mega.nz/#!ZVc0EA7D!zSm7BfGh7EeVW3gkuXdIEQ9sGxna7OtCHt2jsrZrGdE

and here is the preset:
https://mega.nz/#!hRVSBS4K!Ddzhrnq3VkPupTcwpaQ_bpJTeMLNPE4gnKHKOjrR9Q4

Make sure to download the DXO camera profiles for my camera. It should detect this as soon as you load up the RAW file and ask if you want to download it.

The preset goes in (Username) > AppData > Local > DXO > DXO Photolab 3 > Presets
 
Last edited:
Orange Nexus said:
So when you said your tuned 3900X for this particular test of DXO was actually not, right? The 9700 at 5.2 was actually tuned but no mention that it was.
Click to expand...
No, the 3900X was tuned for all tests. The 9700K has not been tuned at all, and is still not tuned. Just running Corsair Vengeance 2666Mhz at stock XMP. You can see this in the CPU-Z shot.
 
Is it supposed to take some time? Applying the RX10 MartinP preset is instantaneous.
 
humbug said:
Ok granted but you're exporting to Lightroom, this is what you said, no? So clearly you do use Lightroom, whatever the reason for the 2 second difference once you're in Lightroom a Zen 2 CPU will give you a faster workflow.

I might look at DXOPhotolab, i don't like banding my Email account around for it to get spammed.
Click to expand...
Brother Humbug, let me say again:
Let me make this abundantly clear, if I do go into Lightroom, Nik Collection or Photoshop there is barely any difference in any recent CPU be it Intel or AMD.

This can be my my workflow depending on the photo:
DX0 Photolab > Lightroom > Nik Collection > Lightroom > Photoshop > Lighroom then save file.

It can also be:
DXO Photolab > save/export to file.

The most time consuming event I have on my workflow is the export from DXO Photolab and mainly because of their quite brilliant Prime noise reduction.

I've been using it since version 1 with my old 3930K and then the export would take about 35secs.
 
You see the little white arrow in the image above you? click on it when the export is done. (On the left of Nik Collection)

amigafan2003 said:
Ah, ok. Just leave settings as default?

Edit:- Took 18s with default option on export.
Click to expand...

That's a lot lower than mine, it can't just be clock speed, given how many cores you have i wonder if it used 8 real cores as opposed to my 4+ SMT given i only have 6 real cores.
 
humbug said:
That's a lot lower than mine, it can't just be clock speed, given how many cores you have i wonder if it used 8 real cores as opposed to my 4+ SMT given i only have 6 real cores.
Click to expand...

Core usage looks the same as yours.

mVGeSSf.png

TR's do have lots of cache that does seem to help in other frequency sensitive apps - such as Photoshop. Maybe that's the case with DXO as well as it's certainly poorly optimised for multi thread.

MartinPrince said:
I've highlighted it here, your time maybe lower.
Click to expand...

Ah thanks, yes shows 18s as well.
 
amigafan2003 said:
Core usage looks the same as yours.

mVGeSSf.png

TR's do have lots of cache that does seem to help in other frequency sensitive apps - such as Photoshop. Maybe that's the case with DXO as well as it's certainly poorly optimised for multi thread.



Ah thanks, yes shows 18s as well.
Click to expand...

A large cache does help in some workloads, quite a lot sometimes as it can store a lot of repetitive branch prediction to reuse without crunching the numbers over and over again. TR3 cache is MASSIVE.

However, every second box in that screen shot is an SMT Thread, so every first box is a real core, if you count them through what you will see is every first box is loaded, other than that jumble in the middle, there are 6 or 7 clear real core loads in that image so it looks like its skipping SMT threads on your CPU and using real ones instead.
If you look at mine it just loads up the first 8 boxes, that's 4 cores + SMT.
With you using 8 real cores your performance from that is going to be much higher than mine, the same would probably be true for an 8 core Zen 2 too.
 
Last edited:
humbug said:
A large cache does help in some workloads, quite a lot sometimes as it can store a lot of repetitive branch prediction to reuse without crunching the numbers over and over again. TR3 cache is MASSIVE.

However, every second box in that screen shot is an SMT Thread, so every first box is a real core, if you count them through what you will see is every first box is loaded, other than that jumble in the middle, there are 6 or 7 clear real core loads in that image so it looks like its skipping SMT threads on your CPU and using real ones instead.
If you look at mine it just loads up the first 8 boxes, that's 4 cores + SMT.
With you using 8 real cores your performance from that is going to be much higher than mine, the same would probably be true for an 8 core Zen 2 too.
Click to expand...
Precisely. This is the problem I've generally had with HT/SMT and why I would never get a 9900K as I find Intel's HT to be a worse implementation than Ryzen's SMT.
Could you run the test a few times to see if there is any variation?
 
You must log in or register to reply here.
Back
Top Bottom