Geo vs DVR's on live view

[framespersecond 0]

sounds interesting. we don't use h.264, we use MPEG 4 ASP. Does ASP encode it diagonally, too? I will try MPEG4 and see, if it'll change.

 

we tried on different PCs: single core, dual core etc. (all intel) - the same problem but only with high speed domes with high speed! nobody have experienced these type of problems?

 

we use also low speed domes - there we don't have any problems

 

I'll try to change the codec and let you know...

[tpolus 0]

sounds interesting. we don't use h.264, we use MPEG 4 ASP. Does ASP encode it diagonally, too? I will try MPEG4 and see, if it'll change.

 

we tried on different PCs: single core, dual core etc. (all intel) - the same problem but only with high speed domes with high speed! nobody have experienced these type of problems?

 

we use also low speed domes - there we don't have any problems

 

I'll try to change the codec and let you know...

 

I also haven't experienced these type of problems. It really seems to be software related. Try to switch to Wavelet, MPEG4 or GeoMPEG4. If it helps, try to reinstall all codecs from C:\GV-xxx\InstallCodec folder.

[framespersecond 0]

I haven't tried your suggestions yet,

 

but I read in the internet, that the effect we have on our screen might be "tearing". I wonder why I have horizontal tearing at the crt screen and diagonal tearing at the TFT screen.

 

what do you think?

[tpolus 0]

I haven't tried your suggestions yet,

 

but I read in the internet, that the effect we have on our screen might be "tearing". I wonder why I have horizontal tearing at the crt screen and diagonal tearing at the TFT screen.

 

what do you think?

 

Please try to record this and send us the link so we could see.

[VJD 0]

 

Currently, I'm testing such stack 2 x GV-2008 cards (also with MultiQuad card). Recording quality is much better than GV-1480 (MPEG2 makes the whole difference but the storage space needed is extremely high - you know, like DVD). CPU usage is of course lower, PC CPU is used only for live view and that's why live view quality is the same as in GV-1480. Personally, didn't notice any quality improvements here. But for perfetly clear live view I really recommend using GV-MultiQuad. It makes the whole difference.

 

So are you saying that the MultiQuad output looks better than the TV out on the Combo Card?

[tpolus 0]

 

Currently, I'm testing such stack 2 x GV-2008 cards (also with MultiQuad card). Recording quality is much better than GV-1480 (MPEG2 makes the whole difference but the storage space needed is extremely high - you know, like DVD). CPU usage is of course lower, PC CPU is used only for live view and that's why live view quality is the same as in GV-1480. Personally, didn't notice any quality improvements here. But for perfetly clear live view I really recommend using GV-MultiQuad. It makes the whole difference.

 

So are you saying that the MultiQuad output looks better than the TV out on the Combo Card?

 

No, MultiQuad output looks pretty much the same as Combo TV output. In my opinion, it's better than GV-Combo's live view because it's not digitized and compressed.

[cctv_down_under 0]

Basically anytime you are watching video on a pC whether through the DSP live display or just the Capture card itself, you are watching digitized video, this will never be as good as analogue video for viewing for many reasons, the first is that there are many video standards, so in order to have capture chips that accomodate all standards, they need to basically crop or stretch the images they also haev issues when interlacing the video.

 

When you dont have the DSP card you are almost watching what you are recording, compressed video and when you use the DSP you are basically watching minimized, stretched and digitized video, the reason standalones dont have this issue is they do not have to digitize the video for display through the video card to your monitor.

 

The reason the DSP is so expensive is that in order to get 25fps real time display you need 16 processing chips (1 for each camera) but if you only recorded 4 cameras without a DSP you could use one or two 25fps chips, IE 25fps divided by the number of cameras, so it is cheaper to manufacture.

 

Yes you can Tpiece off the fly leads, but what a mess, the biggest issue is that the MUXing component, IE getting many cameras on one chip is expensive so if you only wanted to choose 1 camera at a time in clear vision you could use what is known as a looping switcher and you would wire the cams to this and then to your DVR, but then you have to press a button for each camera, you could use a MUX with looping outputs but once again you are paying for something the DVR already does.

 

Looping switchers are very common indeed, but the new combo cards allow for programmabe DSP, this means you can choose any layout from the cameras displayed on a AV Capable montior, the trade of is that you loose the DSP smooth effect on the DVR if you enable it.

 

Basically a looping switch will look 100%, using a the looping function of the DSP will look 85% and converting the video from digital to analogue or visa versa is about 50%

[CollinR 0]

You should not use those BNC "T"s, get something to loop out.

 

There is a little more to it then that, Geo makes the assumption you do not want MPEG2 style storage requirements and furthers it in their codecs. Better codecs exist for similar video however there is no CCTV system I know of that really pushes that.

 

I didn't know about looping switchers, this may be even cheaper and provide good quality results. Another option is the Geo loop out card (not too bad $) and a plain old auto switch. Way way cheaper then the 1480 (vs 800 which I prefer), cheaper then the Geo DSP add-on, cheaper then the multi-quad although not nearly as functional either.