USB traffic, setting for ASI120mc camera, in Sharpcap and Firecapture

[Guest]

Hi all,

I have had the above camera for a while now and has been working great, I was getting around 13.7 fps at maximum resolution, which in think is about the norm for a mid range speed laptop, according to a few other posts I have read.

I have since upgraded to the newer versions of Sharpcap and the new Beta Firecapture software, also,the ASI camera driver, and since then I still get around the same frame rates, but the picture is very flickery and also it is dropping around half of the frames just when previewing.

I have found this new setting that was not there before, well I don't think it was, the USB traffic setting and it's default is set at 80%, I have lowered this to 40% and solved all the issues.

But my question is what was this setting running at by default before the feature was there to adjust, because it ran fine before, so would it have been an automatic setting and would have lowered itself to 40% without the user knowing! and now it has been made adjustable??

And what effect will it have on my imaging having this setting as low as 40%, when the default is 80% ??

Seems a big reduction to me, but through all this my frame rates have not changed, whether USB traffic setting at 40% or 80% the frame rate at maximum res is 13.7fps, but at the higher traffic setting most of the frames are dropped and the picture breaks up every few seconds, at the lower all is fine.

Hope that makes sense, and some could help answer my questions..........

Regards

Olly

Edited September 22, 2014 by Guest

[Pete Presland]

its been a while since i used firecapture, but i think the USB traffic setting has been there for a while, it is certainly on my older version. i normally have mine set around 55-60% high enough to allow a good frame rate, not too high to start dropping frames. 13 fps is pretty slow, what are you imaging?

[D4N]

It sounds like the bottleneck is somewhere else, either your hard disk write speed or the exposure you have set.

I normally have it at 60-70% USB traffic.

Sent from my iPad using Tapatalk

[Pete Presland]

more info needed about your capture settings. are you using the full 1280x960 format or ROI format, what exposure time?

[knobby]

I have mine at 70 % , its to slow the data rate enough to suit your PC. Imagine the PC is a bath, the USB cable a leaky hose pipe and the camera is the tap ... The USB traffic setting is like a limit on how far the tap opens ... Too far and your water (data) starts leaking out if the leaky hosepipe (USB cable)

[Guest]

more info needed about your capture settings. are you using the full 1280x960 format or ROI format, what exposure time?

This was all just for test purposes, I was using full 1280x960 format and auto exposure and no ROI in daytime conditions on preview mode not capture, if I have the USB traffic up any higher than 40 I get dropped frames and any higher than 65 and then the picture starts to flicker every few seconds, put it up to 100 and it freezes the picture all blocky.

But I have never altered that setting before it was on the default 80%, the problems only started after upgrading sharpcap, Firecapture and the ASI driver, it was perfect before on the higher setting

Regards

Olly

[Guest]

One more question, does anybody know if the new USB 3 version gives better frames per second on a USB 2 computer! Or will it work just the same as the USB 2 version......sorry if it is a stupid question...

Olly

[D4N]

I believe it has a larger buffer so should have less dropped frames even on USB 2, whether this is the case in reality I don't know as I don't have one.

You didn't update the firmware did you? Your only supposed to do that if the camera isn't working properly, like if you try to use it on a Mac. You can roll back the firmware with the same program that updates it if you did.

Sent from my iPad using Tapatalk

[stoffe]

I don not think the USB3.0 version performs better then the USB2.0 version on a USB2.0 port. The USB2.0 ASI120MM camera do not like any other devices on the same USB host. I have my USB traffic setting at 96% on my telescope computer and every other USB device is connected on a separate PCI USB2.0 card. The ASI120MM camera is alone on the motherboards USB host. If I connect a USB mouse on any other port to the motherboard I have to lower the USB traffic setting, and yes a mouse shouldn't draw any bandwidth at all but I think it has something with resource allocation to do. I am not an expert on these type of things, btw I have reached 100% USB traffic on a P5Q motherboard!

But on a laptop, there are many internal USB resources that can't be disconnected. But you can try different USB jacks and if you have a PCI express or PC-card slot, a slot in card could be a solution

[JamesF]

I don not think the USB3.0 version performs better then the USB2.0 version on a USB2.0 port. 

I have to differ on this.  I find the USB3 camera is more stable than the USB2 model, even on a USB2 port.

James

[rwg]

Yes, James is right. I talked to Sam at ZWO about this - the USB 2 cameras have an onboard buffer of only 4Kb - if this fills up because the PC does not fetch data quickly enough then you get a dropped frame. The USB3 camera has a 48Kb onboard buffer, so they are much more tolerant of PCs where the USB chipset is a bit iffy. I also got >30fps on USB2 with the USB3 version of the camera - about 33 if I remember correctly.

Not sure why your 'stable' value has changed from 80 to 40%, but if you are getting the same maximum frame rate as before then no harm done I guess. You could try putting SharpCap, the driver and Firecapture back to their old versions (one at a time) to see if you can work out which one it is causes the difference, but I suspect it's not really worth the effort.

cheers,

Robin

[Guest]

I have to differ on this.  I find the USB3 camera is more stable than the USB2 model, even on a USB2 port.

James

Hi James,

I noticed in another thread that you said you could not go much above 40% on the USB traffic setting, and you were getting around 13.7 fps at maximum res on the camera, what computer was that with, as those settings are identical to what I can get, and I use an Acer aspire one netbook with 1.6ghz dual core processor with 2 gb of ram, so would be interested to know your computer spec.

Regards

Olly

[Guest]

Yes, James is right. I talked to Sam at ZWO about this - the USB 2 cameras have an onboard buffer of only 4Kb - if this fills up because the PC does not fetch data quickly enough then you get a dropped frame. The USB3 camera has a 48Kb onboard buffer, so they are much more tolerant of PCs where the USB chipset is a bit iffy. I also got >30fps on USB2 with the USB3 version of the camera - about 33 if I remember correctly.

Not sure why your 'stable' value has changed from 80 to 40%, but if you are getting the same maximum frame rate as before then no harm done I guess. You could try putting SharpCap, the driver and Firecapture back to their old versions (one at a time) to see if you can work out which one it is causes the difference, but I suspect it's not really worth the effort.

cheers,

Robin

Hi Robin,

Thanks for the info

I am getting the same frame rates as before, but I did not know about the traffic setting before so never touched it, so I am assuming that it would have been set at the default of 80% and all worked fine! but since the upgrades I have to lower to around 40 % to stop dropped frames and flickering of the image .... very strange

Regards

Olly

[JamesF]

I can't recall off the top of my head which machine I tested on, but I'm not sure it matters in terms of memory or processor power.  I think it probably has far more to do with the quality of the USB chipset and in particular how well it handles data flow when the bus is close to saturation.  I'd guess it was probably the Acer laptop I use for imaging which is an i3-370 with 3GB RAM, but doesn't have that hot a USB subsystem.

James

[Guest]

I believe it has a larger buffer so should have less dropped frames even on USB 2, whether this is the case in reality I don't know as I don't have one.

You didn't update the firmware did you? Your only supposed to do that if the camera isn't working properly, like if you try to use it on a Mac. You can roll back the firmware with the same program that updates it if you did.

Sent from my iPad using Tapatalk

No have not updated the firmware, although it have considered it, but after what you have said maybe not...

Regards

Olly

[Guest]

Ok, I have done some more tests

In Firecapture I had the 16 bit mode box ticked, so with it unticked and at maximum res with USB setting at 70% I can get 25fps easily, but when debayering is on I get, with same settings, 22fps, so pretty good. But take the USB setting down to 40% which I have to do in sharpcap! then back to 13.7 fps. But it works at 70% which sharpcap won't! although is there anywhere in Firecapture to see how many dropped frames I get though ? They might have all been...!!

So is sharpcap running my camera in 16 bit mode if so how do I turn off, I have tried using the mono8 setting and still only get 17fps and can get 13.7 fps in RGB24 mode, but there are no other bit mode settings, and for all settings I have to use USB traffic at 40% in sharpcap

any higher and problems occur

So all in all it seems that Firecapture handles the camera far far better....... Unless there is something I am missing, I did read somewhere about having the SDK Driver loaded for sharpcap.... Not sure what that meant.

Olly

[AstroAdam]

I've always ended up coming back to 80% USB, and I'm running on a measly 1.6Ghz Atom netbook.  It's an HP/Compaq mini 311, with 2Gb of Ram, running Windows 7, so it must have  agenrally good USB subsystem, as it never really seems to give me any issues.

I get 19fps at full resolution, and if I drop down to a smaller ROI, then depending on the size of that, I get between 35 and 90fps generally using fire capture.  

Oh, and I capture undebayered and let Autostakkert do the debayering afterwards...

Edited September 23, 2014 by AstroAdam

[AstroAdam]

Ok, I have done some more tests

In Firecapture I had the 16 bit mode box ticked, so with it unticked and at maximum res with USB setting at 70% I can get 25fps easily, but when debayering is on I get, with same settings, 22fps, so pretty good. But take the USB setting down to 40% which I have to do in sharpcap! then back to 13.7 fps. But it works at 70% which sharpcap won't! although is there anywhere in Firecapture to see how many dropped frames I get though ? They might have all been...!!

So is sharpcap running my camera in 16 bit mode if so how do I turn off, I have tried using the mono8 setting and still only get 17fps and can get 13.7 fps in RGB24 mode, but there are no other bit mode settings, and for all settings I have to use USB traffic at 40% in sharpcap

any higher and problems occur

So all in all it seems that Firecapture handles the camera far far better....... Unless there is something I am missing, I did read somewhere about having the SDK Driver loaded for sharpcap.... Not sure what that meant.

Olly

I much prefer Firecapture.  I've always found it much more stable, and have got far better results than I ever did with Sharpcap.

Edited September 23, 2014 by AstroAdam

[Guest]

I've always ended up coming back to 80% USB, and I'm running on a measly 1.6Ghz Atom netbook.  It's an HP/Compaq mini 311, with 2Gb of Ram, running Windows 7, so it must have  agenrally good USB subsystem, as it never really seems to give me any issues.

I get 19fps at full resolution, and if I drop down to a smaller ROI, then depending on the size of that, I get between 35 and 90fps generally using fire capture.  

Oh, and I capture undebayered and let Autostakkert do the debayering afterwards...

Use I am using an Acer netbook with 1.6ghz atom dual core, is there anywhere in Firecapture where you can see if any frames are being dropped in the preview mode??

Regards

Olly

[AstroAdam]

I think it basically gets as many frames as it can down the wire per second. The Max fps shows the potential maximum, and the difference between that and the actual fps are the dropped frames. You don't as a such have to worry about dropped frames, as long as your actual fps is sufficient. Dropped frames are a normal part of capture using a camera such as this. That's the difference between using a dedicated planetary cam and a webcam. A webcam will try and maintain a particular frame rate by compressing the video. A dedicated cam just sends as many frames as possible uncompressed, so some will get dropped. The frames you do get however are the best quality possible, even if there are fewer of them.

In this case, ROI is your friend. When you set the ROI, it actually stops the camera sending the stuff outside that area, so you don't get the wasted space clogging up your USB bandwidth, and so FPS goes up.

Edited September 23, 2014 by AstroAdam

[Guest]

I think it basically gets as many frames as it can down the wire per second. The Max fps shows the potential maximum, and the difference between that and the actual fps are the dropped frames. You don't as a such have to worry about dropped frames, as long as your actual fps is sufficient. Dropped frames are a normal part of capture using a camera such as this. That's the difference between using a dedicated planetary cam and a webcam. A webcam will try and maintain a particular frame rate by compressing the video. A dedicated cam just sends as many frames as possible uncompressed, so some will get dropped. The frames you do get however are the best quality possible, even if there are fewer of them.

In this case, ROI is your friend. When you set the ROI, it actually stops the camera sending the stuff outside that area, so you don't get the wasted space clogging up your USB bandwidth, and so FPS goes up.

Hi again

I captured some video earlier with sharpcap and it recorded around 2/3 of the frames and dropped the rest, and when I hit the stop record button it stopped, as you would expect, then I did the same with Firecapture and it recorded, I hit the stop record button and it stopped recording, BUT carried on capturing the frames, playing catch as it had got behind, 15 seconds later it showed that it had captured all the frames, so has it got some sort of buffer that allows this too happen in FC...? If so then superb

Regards

Olly

[riklaunim]

If the HDD isn't fast enough (or busy with other I/O operations) then the frames lag behind cached in RAM.

[stoffe]

There are a speed test option in fire capture maybe you can try that to see if the computer lacks performance? In Windows you can run the disc defragment tool to speed up your hard drive (if you have a mechanical drive).

[AstroAdam]

Remember that capturing debayered video takes three times as much hdd space as undebayered. Therefore capturing undebayered video takes less resources as is not being processed to colour by the cpu on the fly, and the hdd only needs to write at a third of the rate.

Most stacking programs will debayer as part of the processing, so you still end up with a colour end result.

Edited September 24, 2014 by AstroAdam

[Guest]

Remember that capturing debayered video takes three times as much hdd space as undebayered. Therefore capturing undebayered video takes less resources as is not being processed to colour by the cpu on the fly, and the hdd only needs to write at a third of the rate.

Most stacking programs will debayer as part of the processing, so you still end up with a colour end result.

Hi,

Just watched the video you made on your website for using FC

Excellent and very informative

Olly