Jump to content
Tidus79001

Atari VCS FPS Unigine Heaven Benchmark

Recommended Posts

This video shows another Atari VCS owner benchmarking his Atari VCS using Unigine Heaven Benchmark with as averaged 30 FPS. I did the same test on my Atari VCS using the same settings as in his video (Medium quality and 1280x720full screen resolution) but got about 19-20 FPS.  I an to figure out why there is such a huge difference between what he got on his Atari VCS vs what I got on mine. I need to determine why that would be the case.

 

Here are my specs:

- Windows 10 Home
- 32 GB of DDR4 RAM (G.SKILL Ripjaws Series DDR4 PC4-19200 2400MHz F4-2400C16D-32GRS). CPU-Z reports the memory is running in Dual Channel mode at 1197 MHz DRAM Frequency (Uncore Frequency is showing 597.8 MHz, but sometime jumps to 1197.7 if it sits on that tab, but not sure if that mean anything or not)
- 2 TB SAMSUNG 870 QVO SATA III in a USB 3.1 enclosure

I spoke with the author of the video and he said that his specs at the time were as follows:
- Memory was the stock 8gb Kingston (2x4g) that came preinstalled on the Atari VCS
- 500 GB M.2 internal SSD drive (Western Blue 3D NAND WDS500G2B0B)
 
I also have a M.2 installed in my Atari VCS as well as the SATA III SSD that I run the Windows 10 from via a USB 3.1 enclosure, but didn't install Windows on it since I didn't want it to screw with the Atari OS. I'm just a bit hesitant to do this (without knowing this is going to fix the issue or if this what is making my FPS lower) since I am going to have to open my Atari VCS up again and the risk of unwanted changes to the Atari OS boot loader from Windows 10 on the m2.

 

Can anyone tell me what they are seeing for FPS on Unigine Heaven Benchmark running with Medium quality and 1280x720 full screen resolution via a USB 3.1 SATA III drive and also the same or another person post what they are getting from the same test with Window 10 installed on a m.2 drive in their Atari VCS.

 

https://www.youtube.com/watch?v=HAkWAf2Heww

Share this post


Link to post
Share on other sites

Even if only someone running Window 10 via USB 3.1 on their Atari VCS can post their nubmer that would be very helpful to me because it lets me if if the FPS I am seeing with my config is comparable to theirs with that same sort of setup.  The test parameters are run the Unigine Heaven Benchmark using Medium quality and 1280x720 full screen resolution.

Share this post


Link to post
Share on other sites

Ok ran the program on my stock VCS in PC Mode using my current Seagate OneTouch 500gb 3.0 USB SSD. It looks like the program takes a while to build up its max FPS, and therefore its average FPS counter will increase over the first two minutes maybe of the benchmark running. After about a minute I saw a min FPS of 6.9 and Max FPS of 52.1, with an average around 23.7. 

  • Like 1

Share this post


Link to post
Share on other sites

Well, considering that movies are still filmed at 24 FPS, I think that is a fair number.  Broadcasts in 4K are still only done at 30 FPS, not too much more than the movie format that was designated over a century ago.

Share this post


Link to post
Share on other sites

Not quite - 24fps in a videogame looks like shit.  It was bad when the Jag was out in 1993.  It's beyond bad now.  Jeff Minter was trying to achieve 60fps in all of his games from Defender 2000 on - have it in interviews.  In his Defender 2000 stuff, he jokingly called it a 60Hz FUR (Frame Update Rate).

  • Like 3

Share this post


Link to post
Share on other sites
1 hour ago, Stephen said:

Not quite - 24fps in a videogame looks like shit.  It was bad when the Jag was out in 1993.  It's beyond bad now.  Jeff Minter was trying to achieve 60fps in all of his games from Defender 2000 on - have it in interviews.  In his Defender 2000 stuff, he jokingly called it a 60Hz FUR (Frame Update Rate).

Yes, and this is why I want that extra 10 FPS.  I had 32 GB vs the 8 the tested had and he was averaging 10 more FPS than moe

 

The only things I can note is as follow:

- The RAM the author of had was the stock 8 GB of DDR 4 2666MHz (he told me he hadn't upgrade) that came with the console vs the 32 GB of DD4 2400MHz RAM that I have

- The author of the video was using 500 GB M.2 internal SSD drive (Western Blue 3D NAND WDS500G2B0B) vs my 2 TB SAMSUNG 870 QVO SATA III in a USB 3.1 enclosure

 

Just seems odd to me his system getting 10 FPS more on average base of these points:

= 2666MHz vs 2400MHz is negligible in speed different

- My memory size is large enough it aid the GPU

- T\here will not be paging to the SSD so read & write time for USB should me much of a factor (normally once a game is load there isn't going to be constant stream of reading & writing either)

 

Can his negligible different in memory speed or internal m.2 actually be the the difference in 10 FPS.  If that is the case I am OK with adjust my config to match that if mean getting a 10 FPS performance boost.

Share this post


Link to post
Share on other sites
1 minute ago, livingonwheels said:

Overclocked 3200 ram

DSC01308.JPG

Interesting, I probably won't do any overclocking as I don't want to put stress on the console that might cause it damage over time.  That being said can you answer the following?

 

- Are using a m.2 or is it a USB 3.1 Gen 2 SSD you use?

- How did you overclock the RAM (AMD Overdrive or BIOS setting and what settings did you adjust in either of those).?

Share this post


Link to post
Share on other sites
Posted (edited)

This is using Crucial Ballistix 2x8GB DDR4-3200 SODIMM 1.35V
BL2K8G32C16S4B Ram, same bios overclock settings as ETA Prime's video.

 

Edited by livingonwheels

Share this post


Link to post
Share on other sites
2 minutes ago, livingonwheels said:

This is using Crucial Ballistix 2x8GB DDR4-3200 SODIMM 1.35V
BL2K8G32C16S4B Ram, same bios overclock settings as ETA Prime's video.

Can you post the link to that video? I would appreciate it greatly.

Share this post


Link to post
Share on other sites
1 minute ago, livingonwheels said:

Edit: I did remove the cheap thermal pad in the VCS and replaced it with Arctic Silver also.

Yeah, that is smart.  I user Arctic Silver on all my build as that is the best money can buy.  How more stress does the overclocking put on the console.  Is is going to shorten the life of console or overheat it?

Share this post


Link to post
Share on other sites

I did all this to the VCS in February and have had no issues. But to be honest I haven't used it much recently. Hopefully the news Mockduck posted today in this forum comes true, then I'll use it regularly again.  You must use 16gb (8x2), not 32gb (16x2) ram. I have been unable to successfully overclock any brand of 32gb ram in the VCS for some reason. To be safe, use the same ram I used in my description. I tried others and some were unable to be overclocked. 

 

Edited by livingonwheels

Share this post


Link to post
Share on other sites
10 hours ago, Tidus79001 said:

This video shows another Atari VCS owner benchmarking his Atari VCS using Unigine Heaven Benchmark with as averaged 30 FPS. I did the same test on my Atari VCS using the same settings as in his video (Medium quality and 1280x720full screen resolution) but got about 19-20 FPS.  I an to figure out why there is such a huge difference between what he got on his Atari VCS vs what I got on mine. I need to determine why that would be the case.

 

Here are my specs:

- Windows 10 Home
- 32 GB of DDR4 RAM (G.SKILL Ripjaws Series DDR4 PC4-19200 2400MHz F4-2400C16D-32GRS). CPU-Z reports the memory is running in Dual Channel mode at 1197 MHz DRAM Frequency (Uncore Frequency is showing 597.8 MHz, but sometime jumps to 1197.7 if it sits on that tab, but not sure if that mean anything or not)
- 2 TB SAMSUNG 870 QVO SATA III in a USB 3.1 enclosure

I spoke with the author of the video and he said that his specs at the time were as follows:
- Memory was the stock 8gb Kingston (2x4g) that came preinstalled on the Atari VCS
- 500 GB M.2 internal SSD drive (Western Blue 3D NAND WDS500G2B0B)
 
I also have a M.2 installed in my Atari VCS as well as the SATA III SSD that I run the Windows 10 from via a USB 3.1 enclosure, but didn't install Windows on it since I didn't want it to screw with the Atari OS. I'm just a bit hesitant to do this (without knowing this is going to fix the issue or if this what is making my FPS lower) since I am going to have to open my Atari VCS up again and the risk of unwanted changes to the Atari OS boot loader from Windows 10 on the m2.

 

Can anyone tell me what they are seeing for FPS on Unigine Heaven Benchmark running with Medium quality and 1280x720 full screen resolution via a USB 3.1 SATA III drive and also the same or another person post what they are getting from the same test with Window 10 installed on a m.2 drive in their Atari VCS.

 

https://www.youtube.com/watch?v=HAkWAf2Heww

Just noticed that my link for the overclockign video in my orignal post isn't a hyperlink.  I include a proper link here for anyone who wants to check itout

 

https://www.youtube.com/watch?v=HAkWAf2Heww

Share this post


Link to post
Share on other sites

So I bought these Ram sticks to upgrade the VCS, since it's 2400 mhz, is it not going to help game performance or will it still help? I'm not really a PC person.

Screenshot_20210818-172816_Amazon Shopping.jpg

Share this post


Link to post
Share on other sites
1 hour ago, SizzleAtari said:

So I bought these Ram sticks to upgrade the VCS, since it's 2400 mhz, is it not going to help game performance or will it still help? I'm not really a PC person.

If you're not using the VCS as a pc then faster ram won't matter.  You would have to get into the Bios and set the clock speed of the ram faster for it to have any effect (If you had, say, 3200 MHz ram instead).  As far as 16 GB capacity in the VCS, It might make a difference in some games or smooth out frame rates etc.  A worthy upgrade if Atari brings some games with higher graphic requirements to the store.

  • Like 1

Share this post


Link to post
Share on other sites
22 hours ago, livingonwheels said:

I did all this to the VCS in February and have had no issues. But to be honest I haven't used it much recently. Hopefully the news Mockduck posted today in this forum comes true, then I'll use it regularly again.  You must use 16gb (8x2), not 32gb (16x2) ram. I have been unable to successfully overclock any brand of 32gb ram in the VCS for some reason. To be safe, use the same ram I used in my description. I tried others and some were unable to be overclocked. 

 

I have 32GB RAM in mine overclocked with no problems. The RAM is rated for 2666mhz.

Share this post


Link to post
Share on other sites
3 hours ago, joeatari1 said:

If you're not using the VCS as a pc then faster ram won't matter.  You would have to get into the Bios and set the clock speed of the ram faster for it to have any effect (If you had, say, 3200 MHz ram instead).  As far as 16 GB capacity in the VCS, It might make a difference in some games or smooth out frame rates etc.  A worthy upgrade if Atari brings some games with higher graphic requirements to the store.

I use it with Windows 10. I haven't used the Atari Console portion in 3 or 4 months. Was just hoping I wasn't wasting my money by adding Ram if it won't help with game performance. Streets of Rage 4 won't even start, it freezes before the title screen. Was hoping that type of issue could be fixed by adding more Ram.

Share this post


Link to post
Share on other sites

I did get in contact with the author of the video and he tested via USB 3.1 for me to compare that against the results he got with the stock 8 GB with Windows 10 installed in PC mode. Also he now has 16GB as opposed to the 8 GB stock memory that he used during his testing in January. The results of his new test are also 30-32 FPS with 16 GB of memory vs 8 GB stock and USB 3.1 this time vs the m.2 he used previously. Also another person with a stock Atari VCS with no upgrades running on a USB 3.0 SSD saw 23.7 FPS on his test.

 

My take away from this is that the m.2 has little to no impact on the FPS (if there was frequent reading & writing this would would be more of a factor) and looks like my use of 2400 MHz vs the stock 8 GB Kingston that run at 2600 MHz is most likely the major cause of the lower FPS. I went with 32 GB G.Skill DDR4 2400 as was advised by another backer but found that many people mistakenly thought the stock memory was 2400 MHz but I looked up the specs last night on the stock stick and can say for a fact that it is has a 2666 Mhz clock speed. I have now ordered 32 G of DDR4 3200 MHz and will put that in my console. I will give my sister the 32 GB of 2400 MHz memory for the Atari VCS that I am giving to her (she will only run the Atari VCSr or emulators of old classic consoles in PC mode so this lower speed memory won't be a big deal for her).

 

A few safe & useful things to do will following the steps in this video I have linked on this reply is as follows (the default setting in the BIOS are based on the stock 8 GB of DD4 2666 MHz memory so these should be adjusted to reflect the specs of the upgraded memory and allocate the additional memory to the GPU to optimize performance).

-------------------------------------------------

- Set the memory speed in the BIOS to ensure that clock on that is matches the rated speed for the memory.

- Allocate more system RAM as video memory (I'm going to go with 4 GB since I have 32 GB installed in my Atari VCS and that will be a nice boost in games that use a lot of video memory and this is double the default 2 GB that was based on an Atari VCS with the stock 8 GB of memory).

*One other thing shown in the video was how to change the TDP from 35W to 54W but I don't think that I will do that as I don't want to put any undue stress on my Atari VCS and risk damaging or overheating it (for those who want to do this it is advised to replace the stock thermal pad with Arctic Silver 5 thermal paste to enhance the  cooling).

 

 

Share this post


Link to post
Share on other sites
On 8/18/2021 at 11:04 PM, Tidus79001 said:

I did get in contact with the author of the video and he tested via USB 3.1 for me to compare that against the results he got with the stock 8 GB with Windows 10 installed in PC mode. Also he now has 16GB as opposed to the 8 GB stock memory that he used during his testing in January. The results of his new test are also 30-32 FPS with 16 GB of memory vs 8 GB stock and USB 3.1 this time vs the m.2 he used previously. Also another person with a stock Atari VCS with no upgrades running on a USB 3.0 SSD saw 23.7 FPS on his test.

 

My take away from this is that the m.2 has little to no impact on the FPS (if there was frequent reading & writing this would would be more of a factor) and looks like my use of 2400 MHz vs the stock 8 GB Kingston that run at 2600 MHz is most likely the major cause of the lower FPS. I went with 32 GB G.Skill DDR4 2400 as was advised by another backer but found that many people mistakenly thought the stock memory was 2400 MHz but I looked up the specs last night on the stock stick and can say for a fact that it is has a 2666 Mhz clock speed. I have now ordered 32 G of DDR4 3200 MHz and will put that in my console. I will give my sister the 32 GB of 2400 MHz memory for the Atari VCS that I am giving to her (she will only run the Atari VCSr or emulators of old classic consoles in PC mode so this lower speed memory won't be a big deal for her).

 

A few safe & useful things to do will following the steps in this video I have linked on this reply is as follows (the default setting in the BIOS are based on the stock 8 GB of DD4 2666 MHz memory so these should be adjusted to reflect the specs of the upgraded memory and allocate the additional memory to the GPU to optimize performance).

-------------------------------------------------

- Set the memory speed in the BIOS to ensure that clock on that is matches the rated speed for the memory.

- Allocate more system RAM as video memory (I'm going to go with 4 GB since I have 32 GB installed in my Atari VCS and that will be a nice boost in games that use a lot of video memory and this is double the default 2 GB that was based on an Atari VCS with the stock 8 GB of memory).

*One other thing shown in the video was how to change the TDP from 35W to 54W but I don't think that I will do that as I don't want to put any undue stress on my Atari VCS and risk damaging or overheating it (for those who want to do this it is advised to replace the stock thermal pad with Arctic Silver 5 thermal paste to enhance the  cooling).

 

 

Finally got the 3200 MHz of memory and still was only getting 20 FPS on the Unigine Heaven Benchmark (BIOS setting have been adjust for the speed and to allocate 4 GB of RAM to the GPU).  Then I went the Samsung Magician utility for my 2 TB SAMSUNG 870 QVO SATA III that is in USB 3.1 enclosure and enabled Rapid mode.  After enabling Rapid more I ran the Unigine Heaven Benchmark again and now set 31.8 FPS.  Happy now with results are they are in line with yours.  I was surprised to see that Rapid mode had that effect, butt 11.8 FPS more is a huge improvement and I am now set to have the best experience possible with games running in PC mode.

 

That being said another Atari VCS owner that I spoke with said his WD 500 GB Sata III SSD is getting 34 FPS.  How do those who don't have a Samsung SSD get this performance? Rapid mode is a Samsung proprietary feature is my understanding.  I bought an Atari VCS for my sister and she is only get the 20 FPS with the SK Hynix S31 SSD I got for her.  Apparently not all SSD brands are not created equal.  There is no such feature on the SK Hyix drives or other SSD I have used.  I guess WD had it turn on by default in the firmware.  Are there other SSD's that offer this function as as well under another name or so method for enabling something similar?

Share this post


Link to post
Share on other sites

I was mistaken about what I said about the RAPID mode setting for my Samsung SSD that is connected vi USB.  I didn't enable RAPID more on the SATA III as it does support it via USB.  I had accidentally enabled it on the internal Samsung m2 NMVe drive and this would have not bearing on the score. I found when running the bench mark again I was getting about 19-20 FPS on average.

 

I decided to test with another enclosure that was only USB 3.0 (I noted that the author of the video told me the he had used a USB 3.0 enclosure for the benchmark when running PC mode from USB).  When I did that found my speed to be 31 FPS.  Now this was frustrating since my other enclosure is a USB 3.1 Gen 2.

 

This all being said I finally figured out what the issue was.  I was considering replacing my USB 3.1 Gen 2, but then suddenly I realized it was likely the cable,  I thought of this because I had a issue with Android auto where the cable that came with the phone was coking on the data being pushed through it, so I bought high quality USB cable to replace it and that solved the issue with the phone (many people online reported the same thing).  So, I grabbed the cable I bought for the phone and connected my USB 3.1 Gen 2 via it my Atari VCS and ran the test again. Now I am averaging 31.9 with score of 793 (ran it a few times and all is holding steady level of FPS).  All of this because manufacturers sometime use shitty quality USB cables.

 

Just wanted to share all of this with everyone in case anyone else has a similar FPS issue on their Atari VCS.

  • Like 2

Share this post


Link to post
Share on other sites
On 9/13/2021 at 12:34 AM, Tidus79001 said:

I was mistaken about what I said about the RAPID mode setting for my Samsung SSD that is connected vi USB.  I didn't enable RAPID more on the SATA III as it does support it via USB.  I had accidentally enabled it on the internal Samsung m2 NMVe drive and this would have not bearing on the score. I found when running the bench mark again I was getting about 19-20 FPS on average.

 

I decided to test with another enclosure that was only USB 3.0 (I noted that the author of the video told me the he had used a USB 3.0 enclosure for the benchmark when running PC mode from USB).  When I did that found my speed to be 31 FPS.  Now this was frustrating since my other enclosure is a USB 3.1 Gen 2.

 

This all being said I finally figured out what the issue was.  I was considering replacing my USB 3.1 Gen 2, but then suddenly I realized it was likely the cable,  I thought of this because I had a issue with Android auto where the cable that came with the phone was coking on the data being pushed through it, so I bought high quality USB cable to replace it and that solved the issue with the phone (many people online reported the same thing).  So, I grabbed the cable I bought for the phone and connected my USB 3.1 Gen 2 via it my Atari VCS and ran the test again. Now I am averaging 31.9 with score of 793 (ran it a few times and all is holding steady level of FPS).  All of this because manufacturers sometime use shitty quality USB cables.

 

Just wanted to share all of this with everyone in case anyone else has a similar FPS issue on their Atari VCS.

That's a very interesting experience, and yes, it's easy to overlook possibilities like that, because we assume quality.
Well done for your persistence and testing methodology.

  • Like 2

Share this post


Link to post
Share on other sites

I add in regard to usb enclosure make sure you drive uses 1A or less or is powerd.     NVME m.2 for me got slower speed and laptop 2.5" disc...   the lowest powered sata 2.5" got me close to the max USB 3.1 speed with is still slower than sata but fast enough for this little PC

Share this post


Link to post
Share on other sites
On 9/23/2021 at 12:02 PM, haightc said:

I add in regard to usb enclosure make sure you drive uses 1A or less or is powerd.     NVME m.2 for me got slower speed and laptop 2.5" disc...   the lowest powered sata 2.5" got me close to the max USB 3.1 speed with is still slower than sata but fast enough for this little PC

Mine is a powered USB 3.1 enclosure and even my USB 3.0 non power enclosure was getting better speeds until I changed to a high grade USB cable on the powered USB 3.1 enclosure.  This was the best rated enclosure on Newegg with a 5 star rating so it was a bit shocking to find out it was performing so badly and really annoying that I found the manufacture is using shoddy cables that robs the performance.  I built another Atari VCS for my sister with the same model powered USB 3.1 enclosure and had the same issue before swapping the USB cable for a high grade on the same as what I changed to on mine.

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...