Show Posts

136

General / Disable Core for GPU

« on: March 12, 2012, 01:35:55 AM »

I have been wondering if you have a CPU with hyper threading do you disable 1 or 2 cores in Photoscan, due to Hyper threading being virtual core.

So for example if you have the i7 3930K it has 6+6 "cores" thats how Photoscan sees it.

I got a GTX 580 and I was wondering if I should disable 1 or 2 cores in Photoscan. Is there a thread enough to feed the GPU or does it need the whole real core for itself.

137

General / Re: Using Chunks workflow...

« on: March 03, 2012, 02:21:26 PM »

Just a little question on the side, I have been wondering regarding using full res images for reconstruction if that is indeed needed, or there to say getting the best results.

Reason I am bringing this up is regarding the bayer sensor as pixel at full resolution are being created, therefore 1/3 of the full resolution should be in theory enough detail for the reconstruction and should get roughly the same amount of accuracy as the full resolution.

This of course does not count the texturing step into this as you would always want to use the full res for the texturing part.

Just a little side thought I have been thinking about ever since I started in photogrammetry, and my fascination by the foveon sensor. (I have not done comparison between the bayer and faveon, but would love to see some tests at full res)

Again just a little side note that could maybe in theory save a lot of processing time.

As well, if you are indeed running out of RAM with 96GB I would take into consideration to get something like this for RAM cache.
http://www.ocztechnology.com/products/solid-state-drives/pci-express/workstation.html

138

General / Re: Graphics cards and memory

« on: March 02, 2012, 07:38:59 PM »

Quote from: Kjellis85 on March 02, 2012, 03:16:56 PM

Quote from: olihar on March 02, 2012, 02:13:25 PM
Any more info on this, I am trying to decide if to take 1.5GB or 3.0GB GTX 580 card.

Unless you have an urgent need for it I suggest waiting for the 680 release in april (or at least that is what the rumors say). http://lenzfire.com/2012/02/entire-nvidia-kepler-series-specifications-price-release-date-43823/

Yes I have been keeping an eye on this, I will then probably take the 1.5 as its cheaper...

Can get this one for a pretty good deal.
http://www.msi.com/product/vga/N580GTX-TWIN-FROZR-III-15D5-POWER-EDITION-OC.html

P.s. the rumored 690 looks pretty good, 2x1024 Cuda cores... pretty insane. 1000$ price to go along with it.

139

General / Re: Graphics cards and memory

« on: March 02, 2012, 02:13:25 PM »

Any more info on this, I am trying to decide if to take 1.5GB or 3.0GB GTX 580 card.

140

General / Graphics cards and memory

« on: February 25, 2012, 12:35:29 PM »

Is there any gain for Photoscan to have a card with 3.0GB RAM instead of 1.5GB

For example you can get the GTX580 in 1.5 and 3.0 GB flavors

Often the 1.5 cards are clocked higher then the 3.0 cards so that would impact this as well.

What is the gain if any?

141

General / Re: Generating mesh - cancel

« on: August 25, 2011, 06:52:44 PM »

Yes as I stated I can do that, but that will result in Photoscan not clearing the cache used on Harddrive.

142

General / Generating mesh - cancel

« on: August 25, 2011, 05:33:56 PM »

Sometimes if I cancel the Generating mesh stage, it can take more then 1 hour for it to cancel.

I can kill the app but that will result in not clearing the cache it has been using.

143

General / Nvidia Tesla and Photoscan

« on: August 16, 2011, 03:03:11 PM »

Can Photoscan use Nvidia Tesla cards, or is the missing shaders a problem for it.

Say for example the NVIDIA Tesla “Fermi” M2050 GPUs

144

Bug Reports / Re: Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 06:07:28 PM »

Sent link to the file in email, would be great to get to know when support for EPSG::3057 is there, I guess adding EPSG::5325 would also be great, as its the same just newer.

Thanks.

145

Bug Reports / Re: Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 05:28:22 PM »

I got a test file with rough 3 ground control points, meant for then EPSG::3057...

Will it help to send that one, or will the EPSG::3057 be no problem to add?

146

Bug Reports / Re: Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 05:21:18 PM »

I just found out, I am using the EPSG::3057

So:
ISN93 / Lambert 1993

That would be the most used one, 2004 is slowly coming through. Then i guess it would be EPSG::5325

147

Bug Reports / Re: Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 04:46:01 PM »

I can see Photoscan has only EPSG::4659

I wonder if I am using EPSG::4944 or some other...

Let me look into this a little bit.

There are 12 ISN in EPSG...

148

General / Re: Agisoft PhotoScan 0.8.3 pre-release

« on: August 12, 2011, 04:00:26 PM »

Reason was I wanted to check if I was getting the same errors in Ground Control in the older versions. But then I ran a test in the older versions and got the same problem, I then posted it in the bug forum about the Datum.

149

Bug Reports / Re: Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 03:58:13 PM »

Quote from: Alexey Pasumansky on August 12, 2011, 01:14:46 PM

Hello olihar,

You need to input data in the coordinate system you are planning to work with.
PhotoScan supports conversion between coordinate systems in the same Datum, but conversion between different Datums is not supported by now.

Are you sure that data on "ISN93" screenshot is in real coordinates but not in local?

Yes thats what I did first inserted ISN93 data into the fields and used ISN93 and got this massive error, then I tried same data using local and it seemed fine.

Then I manually converted the ISN93 data into WSG84 as I wanted to try and see what Photoscan would do then. Then the identical location data just different Datum/format and it worked fine as you can see.

I just wonder if there is something wrong with the ISN93 in Photoscan.

150

Bug Reports / Ground Control and Coordinate system - crazy error.

« on: August 12, 2011, 12:17:38 PM »

Seems very strange to me, the only coordinate system I seem to get to work is local Coordinates and WGS 84, I started with ISN93 and converted them straight to WGS 84 so its the same location.

Local Coordinates, looks good.