Photoshop faster on 8700K without Hyperthreading

[gipper51]

Playing around with my PC today, and ran the Puget Systems 2018 Photoshop benchmark with some different settings. Interestingly enough, Photoshop CC 2018 is running faster with hyperthreading turned off! This included the benchmark scores and also a test of exporting a batch of RAW files to JPG. I"m surprised at the export times as it uses all cores when doing so, I'd have thought HT would help here...but it doesn't.

My specs that matter for the test:

8700K overclocked to 4.8Ghz

32GB DDR3200

960 EVO boot drive

1080 GTX

My export test was exporting 90 RAW files from a 7D2 to JPG in ACR. Times to export were as follows.

4.8Ghz With HT: 2:08
4.8Ghz Without HT: 1:58
5.1Ghz Without HT: 1:51

The CPU temps dropped a few degrees without HT, so I cranked it up to 5.1Ghz and tested it. It can run a bit faster with HT off. 128 seconds vs 111 seconds for the export. That's 15% faster.

These were my Puget benchmarks with those settings. Starting at 4.8 with HT, then no HT, then 5.1Ghz with no HT. Overall score went from 1003 to 1044. The temps were monitored in RealTempGT. Cores never exceeded 67C during the test.


4.8 Ghz with hyperthreading on


4.8 Ghz with hyperthreading off


5.1 Ghz with hyperthreading off

I guess I'll leave hyperthreading off from now on. Photoshop is the only program I use that really uses the CPU horsepower, and if it's running faster without HT....

Damn. I know PS is not a program that utilizes multiple threads well, but I didn't expect there to be a negative performance hit with having HT on.



--
My site:

http://www.gipperich-photography.com

 

[fishy wishy]

I wouldn't worry about it, it just means Adobe stuff is not well parallelised, which is something we all knew. The 1% lower score with hyperthreading on presumably reflects the overhead of it trying to divide the work up.

I'm not really that fussed about hyperthreading for this. I mainly like it there for hoping that virtual machines won't hog my physical processors when they're not doing anything.

 

[gipper51]

I wouldn't worry about it, it just means Adobe stuff is not well parallelised, which is something we all knew. The 1% lower score with hyperthreading on presumably reflects the overhead of it trying to divide the work up.

I'm not really that fussed about hyperthreading for this. I mainly like it there for hoping that virtual machines won't hog my physical processors when they're not doing anything.

True. I think I'm going to leave HT off for now and run the machine at 5.1Ghz. In my simple mind I like saying I have "5.1Ghz" instead of 4.8

I have several overclock profiles saved in the BIOS. A reboot and a few seconds in the BIOS I can have hyperthreading back on when needed...which won't be often.

 

[jfriend00]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

--
John

 

[DVT80111]

Playing around with my PC today, and ran the Puget Systems 2018 Photoshop benchmark with some different settings. Interestingly enough, Photoshop CC 2018 is running faster with hyperthreading turned off! This included the benchmark scores and also a test of exporting a batch of RAW files to JPG. I"m surprised at the export times as it uses all cores when doing so, I'd have thought HT would help here...but it doesn't.

My specs that matter for the test:

8700K overclocked to 4.8Ghz

32GB DDR3200

960 EVO boot drive

1080 GTX

My export test was exporting 90 RAW files from a 7D2 to JPG in ACR. Times to export were as follows.

4.8Ghz With HT: 2:08
4.8Ghz Without HT: 1:58
5.1Ghz Without HT: 1:51

The CPU temps dropped a few degrees without HT, so I cranked it up to 5.1Ghz and tested it. It can run a bit faster with HT off. 128 seconds vs 111 seconds for the export. That's 15% faster.

These were my Puget benchmarks with those settings. Starting at 4.8 with HT, then no HT, then 5.1Ghz with no HT. Overall score went from 1003 to 1044. The temps were monitored in RealTempGT. Cores never exceeded 67C during the test.


4.8 Ghz with hyperthreading on


4.8 Ghz with hyperthreading off


5.1 Ghz with hyperthreading off

I guess I'll leave hyperthreading off from now on. Photoshop is the only program I use that really uses the CPU horsepower, and if it's running faster without HT....

Damn. I know PS is not a program that utilizes multiple threads well, but I didn't expect there to be a negative performance hit with having HT on.

1.5% difference, I would bother to fiddle with it.

Adobe don't use more than 8 cores anyway.

 

[jfriend00]

1.5% difference, I would bother to fiddle with it.

Adobe don't use more than 8 cores anyway.

You have a weird way of doing math. Might try calculating again.

 

[fishy wishy]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

If you don't have hyperthreading, won't a virtual machine simply take over the physical cores specified for it? I would have thought that when a VM was doing nothing, with hyperthreading the host OS can still make use of those cores.

 

[jfriend00]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

If you don't have hyperthreading, won't a virtual machine simply take over the physical cores specified for it? I would have thought that when a VM was doing nothing, with hyperthreading the host OS can still make use of those cores.

What does performance in a VM have to do with this conversation? Hyperthreads are just hardware assisted threads. They still share cores and caches within the CPU. Sometimes they are more efficient than OS-level threads, sometimes not.

And, you can still get hyperthreads in some of Intel's other processors, particularly the ones meant for high scale like the Xeon line.

 

[gipper51]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

Yep, I only built the machine a few months ago, and the plan was to upgrade to the 9900K and sell the 8700K. After reviews showed what a hot and expensive power hog that chip is, I decided to pass. The 9700K may still get dropped into this system though. 8 cores at 5GHz+ is awfully tempting. I don't need it, but I'm a geek.

Hyperthreading has shown to be useful for many tasks, but Photoshop certainly doesn't appear to utilize it.

 

[Austinian]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

Yep, I only built the machine a few months ago, and the plan was to upgrade to the 9900K and sell the 8700K. After reviews showed what a hot and expensive power hog that chip is, I decided to pass. The 9700K may still get dropped into this system though. 8 cores at 5GHz+ is awfully tempting. I don't need it, but I'm a geek.

Hyperthreading has shown to be useful for many tasks, but Photoshop certainly doesn't appear to utilize it.

Hyperthreading also enables certain security exploits, but it does appear to benefit efficiently multi-threaded programs like DxO PhotoLab.

So, as usual, various tradeoffs to be considered. :-|

Loading…

arstechnica.com

 

[fishy wishy]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

If you don't have hyperthreading, won't a virtual machine simply take over the physical cores specified for it? I would have thought that when a VM was doing nothing, with hyperthreading the host OS can still make use of those cores.

What does performance in a VM have to do with this conversation? Hyperthreads are just hardware assisted threads. They still share cores and caches within the CPU. Sometimes they are more efficient than OS-level threads, sometimes not.

If you actually READ the question, you'll see the question is re whether a virtual machine on a computer without hyperthreading will hog the cores more than it needs when the VM is doing nothing.

It's not about whether we know that hyperthreading is virtual cores.

And, you can still get hyperthreads in some of Intel's other processors, particularly the ones meant for high scale like the Xeon line.

I have a Xeon and enable hyperthreading. And I don't flame others unnecessarily.

 

[jfriend00]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

If you don't have hyperthreading, won't a virtual machine simply take over the physical cores specified for it? I would have thought that when a VM was doing nothing, with hyperthreading the host OS can still make use of those cores.

What does performance in a VM have to do with this conversation? Hyperthreads are just hardware assisted threads. They still share cores and caches within the CPU. Sometimes they are more efficient than OS-level threads, sometimes not.

If you actually READ the question, you'll see the question is re whether a virtual machine on a computer without hyperthreading will hog the cores more than it needs when the VM is doing nothing.

It's not about whether we know that hyperthreading is virtual cores.

And, you can still get hyperthreads in some of Intel's other processors, particularly the ones meant for high scale like the Xeon line.

I have a Xeon and enable hyperthreading. And I don't flame others unnecessarily.

I just don't understand how your question about VMs (which was addressed to me) is on topic for this conversation or why it is addressed to me. I still don't understand that. No flame intended. I just don't know how it fits with what we are discussing.

 

[jfriend00]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

Yep, I only built the machine a few months ago, and the plan was to upgrade to the 9900K and sell the 8700K. After reviews showed what a hot and expensive power hog that chip is, I decided to pass. The 9700K may still get dropped into this system though. 8 cores at 5GHz+ is awfully tempting. I don't need it, but I'm a geek.

Hyperthreading has shown to be useful for many tasks, but Photoshop certainly doesn't appear to utilize it.

In a single user computer, it also seems like for hyperthreading on an 8-core CPU (like the 9700k) to even have a chance at being helpful, you'd have to run an app that was trying to do significant computation on more than 8 threads because with 8 or less threads active at once, you've already got a whole core just for each thread.

Multi-user servers with lots of active threads are a different situation entirely and that's where the needs of a workstation may be different than the needs of a server.

Since the i7-9700k now has 8 cores (up from 6 in the 8700k), perhaps the dropping of hyperthreads from the i7-9700k is a repositioning of that chip as a desktop chip, not a server chip and that may actually work to our a advantage as desktop users.

 

[jfriend00]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

Yep, I only built the machine a few months ago, and the plan was to upgrade to the 9900K and sell the 8700K. After reviews showed what a hot and expensive power hog that chip is, I decided to pass. The 9700K may still get dropped into this system though. 8 cores at 5GHz+ is awfully tempting. I don't need it, but I'm a geek.

Hyperthreading has shown to be useful for many tasks, but Photoshop certainly doesn't appear to utilize it.

Hyperthreading also enables certain security exploits, but it does appear to benefit efficiently multi-threaded programs like DxO PhotoLab.

So, as usual, various tradeoffs to be considered. :-|

https://arstechnica.com/information...erthreading-exploit-that-pilfers-crypto-keys/

Are there any articles that document hyperthreads used with DxO? I'd like to read.

Note to other readers. The exploit in that link appears to be aimed mostly at infrastructure providers that have multiple tenants on the same CPU chip where one tenant might be able to steal stuff from another tenant on the same CPU via hyperthreading. I did not see how that particular case is a vulnerability we need to worry about for desktop usage. One of the difficulties in all these recent CPU exploits is figuring who they really affect and which ones you need to worry about or not when you're doing desktop computing.

A quote from that article: PortSmash currently poses a threat mainly to people using computers or services that allow untrusted people to use the same physical processor.

 

[Austinian]

Very interesting... And, the successor i7-9700k chip doesn't even have hyperthreading (it has 8 real cores). So, maybe this hyperthreading stuff wasn't really working anyway.

Plus your overclock to 5.1GHz is going to be faster on ALL the single threaded things that Photoshop does too.

Yep, I only built the machine a few months ago, and the plan was to upgrade to the 9900K and sell the 8700K. After reviews showed what a hot and expensive power hog that chip is, I decided to pass. The 9700K may still get dropped into this system though. 8 cores at 5GHz+ is awfully tempting. I don't need it, but I'm a geek.

Hyperthreading has shown to be useful for many tasks, but Photoshop certainly doesn't appear to utilize it.

Hyperthreading also enables certain security exploits, but it does appear to benefit efficiently multi-threaded programs like DxO PhotoLab.

So, as usual, various tradeoffs to be considered. :-|

https://arstechnica.com/information...erthreading-exploit-that-pilfers-crypto-keys/

Are there any articles that document hyperthreads used with DxO? I'd like to read.

Note to other readers. The exploit in that link appears to be aimed mostly at infrastructure providers that have multiple tenants on the same CPU chip where one tenant might be able to steal stuff from another tenant on the same CPU via hyperthreading. I did not see how that particular case is a vulnerability we need to worry about for desktop usage. One of the difficulties in all these recent CPU exploits is figuring who they really affect and which ones you need to worry about or not when you're doing desktop computing.

A quote from that article: PortSmash currently poses a threat mainly to people using computers or services that allow untrusted people to use the same physical processor.

I'm not personally worried about most of the recent CPU side-channel exploits, but I do think it's interesting that they exist, and that complexity can create hard-to-find problems.

As for the multithreading of PhotoLab, here's a screenshot of Process Explorer while I was batch-processing multiple raw files using Prime NR:


Looks like it's making the CPU work pretty hard.

 

[jfriend00]

As for the multithreading of PhotoLab, here's a screenshot of Process Explorer while I was batch-processing multiple raw files using Prime NR:


Looks like it's making the CPU work pretty hard.

Batch processing lots of images is the easiest way to use multi-cores. Even current versions of Lightroom employ multiple cores for exporting. Engaging all the hyperthreads is different than whether they are actually making things run faster or not. One would have to compare the performance of some operations with hyper threads enabled and then without on the same system to judge their effectiveness for that operation.

--
John

 

[Austinian]

As for the multithreading of PhotoLab, here's a screenshot of Process Explorer while I was batch-processing multiple raw files using Prime NR:


Looks like it's making the CPU work pretty hard.

Batch processing lots of images is the easiest way to use multi-cores. Even current versions of Lightroom employ multiple cores for exporting. Engaging all the hyperthreads is different than whether they are actually making things run faster or not. One would have to compare the performance of some operations with hyper threads enabled and then without on the same system to judge their effectiveness for that operation.

Interesting question, so I went looking for data. Didn't seen anything about hyperthreading on vs. off in PhotoLab specifically. :-(

I'm inclined to believe what Process Explorer appears to be showing, that all 12 threads are active and working in PhotoLab, because each thread shows a slightly different usage profile.

But if anyone has data showing that processing times are the same with and without hyperthreading enabled I'd be interested to see it.

I'm on my gaming boot drive right now; if I get curious enough I might do some tests tomorrow on the primary boot drive that has PhotoLab, with HT on vs. off.

 

[gipper51]

As for the multithreading of PhotoLab, here's a screenshot of Process Explorer while I was batch-processing multiple raw files using Prime NR:


Looks like it's making the CPU work pretty hard.

Batch processing lots of images is the easiest way to use multi-cores. Even current versions of Lightroom employ multiple cores for exporting. Engaging all the hyperthreads is different than whether they are actually making things run faster or not. One would have to compare the performance of some operations with hyper threads enabled and then without on the same system to judge their effectiveness for that operation.

Interesting question, so I went looking for data. Didn't seen anything about hyperthreading on vs. off in PhotoLab specifically. :-(

I'm inclined to believe what Process Explorer appears to be showing, that all 12 threads are active and working in PhotoLab, because each thread shows a slightly different usage profile.

But if anyone has data showing that processing times are the same with and without hyperthreading enabled I'd be interested to see it.

I'm on my gaming boot drive right now; if I get curious enough I might do some tests tomorrow on the primary boot drive that has PhotoLab, with HT on vs. off.

I don't have any data other than my own anecdotal test. I can tell you that when I was exporting RAW to JPG in ACR, the task manager was showing all 12 threads full and the CPU usage was at 98-100% for the export with HT on. With HT off it was also pegging all 6 cores to the max. Wish I would have took some screen shots.

But in the end, the process was faster with just the 6 cores. So HT appears to be used for image exporting, but it isn't helping in ACR.

--
My site:

http://www.gipperich-photography.com

 

[jfriend00]

As for the multithreading of PhotoLab, here's a screenshot of Process Explorer while I was batch-processing multiple raw files using Prime NR:


Looks like it's making the CPU work pretty hard.

Batch processing lots of images is the easiest way to use multi-cores. Even current versions of Lightroom employ multiple cores for exporting. Engaging all the hyperthreads is different than whether they are actually making things run faster or not. One would have to compare the performance of some operations with hyper threads enabled and then without on the same system to judge their effectiveness for that operation.

Interesting question, so I went looking for data. Didn't seen anything about hyperthreading on vs. off in PhotoLab specifically. :-(

I'm inclined to believe what Process Explorer appears to be showing, that all 12 threads are active and working in PhotoLab, because each thread shows a slightly different usage profile.

But if anyone has data showing that processing times are the same with and without hyperthreading enabled I'd be interested to see it.

I'm on my gaming boot drive right now; if I get curious enough I might do some tests tomorrow on the primary boot drive that has PhotoLab, with HT on vs. off.

I'd be curious what you find. From what I've read about Hyperthreads, there are some situations where they improve performance a bit and some situations where their extra overhead actually decreases performance a bit. It is very, very application-specific.

I found it very interesting that Intel removed HT from the 8-core i7-9700k when compared to its predecessor, the 6-core i7-8700k and yet the 8-core i9-9900k has HT again. Odd. Maybe trying to position the i9-9900k as their server chip (multi-user/multi-task servers are probably more likely to benefit from HT) and the i7-9700k as a desktop chip. Not sure.

John

 

[gipper51]

In a single user computer, it also seems like for hyperthreading on an 8-core CPU (like the 9700k) to even have a chance at being helpful, you'd have to run an app that was trying to do significant computation on more than 8 threads because with 8 or less threads active at once, you've already got a whole core just for each thread.

Multi-user servers with lots of active threads are a different situation entirely and that's where the needs of a workstation may be different than the needs of a server.

Since the i7-9700k now has 8 cores (up from 6 in the 8700k), perhaps the dropping of hyperthreads from the i7-9700k is a repositioning of that chip as a desktop chip, not a server chip and that may actually work to our a advantage as desktop users.

I don't know if Intel is leaning towards "hyperthreading=workstation" chips or not. Many of the lower end CPUs have HT but they are only dual cores. HT is across the lineup from bottom to top, would be hard to market that as a "premium" feature when older, low end i3 chips have it, yet a new high end i7 doesn't.

Almost seems like Intel is trying to keep everything within a "thread range" on consumer level parts. All the mainstream chips now have between 4-8 threads (dual core with HT, 4, 6, and 8 core without HT) until you get to at least 9900K level and above with the X299 platform. Want more than 8 threads on Intel? Gotta really pay up for it.

That or Intel is keeping HT on more chips in their back pocket in case the next gen Ryzen parts really up the ante. Intel can switch back to HT and gain some performance back if needed. Competition can make for funny decisions sometimes. Overall it's great for us though! You could argue that Intel has made more progress on generation 8/9 alone than they did on 2 through 7, and it's thanks to AMD.

--
My site:
http://www.gipperich-photography.com