How do I import good quality images to MSP8?
-
David_S
- Posts: 47
- Joined: Thu Mar 24, 2005 8:32 pm
- System_Drive: C
- 32bit or 64bit: 64 Bit
- motherboard: Asus P5Q Pro
- processor: Intel Core TM 2 Quad CPU Q6600 2.4GHz
- ram: 4Gb
- Video Card: nVideo GeForce 8800 GTX
- Hard_Drive_Capacity: 4Tb
- Monitor/Display Make & Model: Philips 190B
- Location: New Zealand
How do I import good quality images to MSP8?
I have been trying for some time to import images to a MSP project but am disappointed by the quality - a general fuzziness and lack of sharpness and a rather contrasty image. But when I import a Ulead sample image from the media library the results are just fine. These Ulead pictures are typically jpeg files with Image Attributes of RGB True Color (24 bit), 720x480, 118 pixels/cm resolution Size 1Mb and a file size of 100Kb according to image properties in Photo Explorer.
The usual source for my images is a 4MB digital camera which typically produces Image Attributes of RGB True Color (24 bit), 2272 x 1704 pixels, resolution 28 pixels/cm, Size 11Mb and a file size of 1Mb.
The main difference between Ulead pictures and mine is the resolution, 118 v 28 pixels/cm. What is the significance (if any) of this and why does the larger file have a smaller resolution? My understanding is that it just changes the image size without changing the file size. Is that right?
My project settings in MSP are
PAL (25 fps)
Video tracks: 7
Microsoft AVI files
24 bits, 720 x 576, 4:3, 25 fps
Lower Field First
DV Video Encoder -- type 1
DV Audio -- PAL, 48.000 kHz, 16 Bit, Stereo
Using PhotoImpact XL I have tried changing the resolution of my pictures from 28 to 118 pixels/cm. and also the image size from 2272 x 1704 to 768 x 576 to approximate my MSP settings (PAL). Neither results in an improvement to the image in MSP. My best results so far have been with pictures taken with an old Nikon F3 35mm film camera which were subsequently digitised but these are still not as good as the Ulead samples.
Anyone got any suggestions? By the way, I have remembered to tick the "keep aspect ratio" box in the import picture dialogue box.
David
The usual source for my images is a 4MB digital camera which typically produces Image Attributes of RGB True Color (24 bit), 2272 x 1704 pixels, resolution 28 pixels/cm, Size 11Mb and a file size of 1Mb.
The main difference between Ulead pictures and mine is the resolution, 118 v 28 pixels/cm. What is the significance (if any) of this and why does the larger file have a smaller resolution? My understanding is that it just changes the image size without changing the file size. Is that right?
My project settings in MSP are
PAL (25 fps)
Video tracks: 7
Microsoft AVI files
24 bits, 720 x 576, 4:3, 25 fps
Lower Field First
DV Video Encoder -- type 1
DV Audio -- PAL, 48.000 kHz, 16 Bit, Stereo
Using PhotoImpact XL I have tried changing the resolution of my pictures from 28 to 118 pixels/cm. and also the image size from 2272 x 1704 to 768 x 576 to approximate my MSP settings (PAL). Neither results in an improvement to the image in MSP. My best results so far have been with pictures taken with an old Nikon F3 35mm film camera which were subsequently digitised but these are still not as good as the Ulead samples.
Anyone got any suggestions? By the way, I have remembered to tick the "keep aspect ratio" box in the import picture dialogue box.
David
-
heinz-oz
Regardless of what you heared or think about it, the fact of the matter is this: Whether your images are listed as 118 or 26 px/cm has NO effect on how the images look on a screen. Be it TV or your PC monitor. For any monitor, a pixel is pixel and will occupy one pixel on the screen.
The sample images look good because they are NTSC format DVD frame size. MSP does have nothing to do to these images to resize them.
Your images from your digital camera are too large. They do not fit into the number of pixels you can display on your TV. MSP needs to resize them to the NTSC or PAL frame size. It can only do that by discarding pixels and rendering the adjacent ones to roughly the content it thinks they should have.
That's it in a nutshell. The higher the MP of your digicam, the worse they might look. A 1 MP image looks better on the TV than a 10 MP one. The difference you will only see in print not on a monitor.
The sample images look good because they are NTSC format DVD frame size. MSP does have nothing to do to these images to resize them.
Your images from your digital camera are too large. They do not fit into the number of pixels you can display on your TV. MSP needs to resize them to the NTSC or PAL frame size. It can only do that by discarding pixels and rendering the adjacent ones to roughly the content it thinks they should have.
That's it in a nutshell. The higher the MP of your digicam, the worse they might look. A 1 MP image looks better on the TV than a 10 MP one. The difference you will only see in print not on a monitor.
-
heinz-oz
Sorry, got carried away because I tend to answer the same question about once or twice every week. I forgot to, directly, answer your question even though the answer is given in between the lines.
Resize your images to the right aspect ratio for your TV standard, either 4:3 or 16:9. Don't worry that the pixel count is not exactly the same as the TV frame size. Video pixels are rectangular, photo pixels are square (just take my word for it)
In PI, I resize mine usually to 1200 x 900 but 800 x 600 or, because I live in PAL land, 720 x 576 (PAL frame size) will do the same. I also save the images to bmp rather than compressing them again. Every time you save a jpg image you apply the same compression again, losing quality.
They come out as clear as I expect to see them.
Resize your images to the right aspect ratio for your TV standard, either 4:3 or 16:9. Don't worry that the pixel count is not exactly the same as the TV frame size. Video pixels are rectangular, photo pixels are square (just take my word for it)
In PI, I resize mine usually to 1200 x 900 but 800 x 600 or, because I live in PAL land, 720 x 576 (PAL frame size) will do the same. I also save the images to bmp rather than compressing them again. Every time you save a jpg image you apply the same compression again, losing quality.
They come out as clear as I expect to see them.
-
pfarrellop
images and resolution
I was glad to find this post because it answers my question. Now if I can only figure out the how of the answer
I have MediaStudio 8, and I use Adobe Photoshop CS2 for working with my images. How do I resize my images to 4:3 for TV? And would psd's work as well as bmp's? There's no compression with that file type either.
Thanks, Pat
I have MediaStudio 8, and I use Adobe Photoshop CS2 for working with my images. How do I resize my images to 4:3 for TV? And would psd's work as well as bmp's? There's no compression with that file type either.
Thanks, Pat
-
sjj1805
- Posts: 14383
- Joined: Wed Jan 26, 2005 7:20 am
- System_Drive: C
- 32bit or 64bit: 32 Bit
- motherboard: Equium P200-178
- processor: Intel Pentium Dual-Core Processor T2080
- ram: 2 GB
- Video Card: Intel 945 Express
- sound_card: Intel GMA 950
- Hard_Drive_Capacity: 1160 GB
- Location: Birmingham UK
MediaStudio will accept the PSD format and so if that's the format they are now in there is no need to convert them to BMP
Please view:
What dimensions etc should I use for still images in Videos?
As for resizing them with CS then the easy way is to create a PhotoShop 'Action' and then use the batch option
File | Automate | Batch
Please view:
What dimensions etc should I use for still images in Videos?
As for resizing them with CS then the easy way is to create a PhotoShop 'Action' and then use the batch option
File | Automate | Batch
-
heinz-oz
Re: images and resolution
If your images are not 4:3 aspect to start with, you will need to crop them. 4:3 means that, if you divide the pixel count along the width of your image by 4 and multiply the result of that division by 3 for the hight of your image.pfarrellop wrote:I was glad to find this post because it answers my question. Now if I can only figure out the how of the answer![]()
I have MediaStudio 8, and I use Adobe Photoshop CS2 for working with my images. How do I resize my images to 4:3 for TV? And would psd's work as well as bmp's? There's no compression with that file type either.
Thanks, Pat
It doesn't matter what format the images are in. If the frame size is correct and the image is in jpeg, fine. Only if you have an image which needs to be modified i.e. resized, cropped, lightened, darkened, rotated etc. it needs to be saved again afterwards. If you save it as jpeg again you'll lose quality. Better to use a format which does not apply another compression to it.
Hi Guys
I was about to pose a similar question to David's as my images look a bit poorand I try and avoid if possible. In photoshop there's a video filter offering the option to de-interlace - should that be used?
I was about to pose a similar question to David's as my images look a bit poorand I try and avoid if possible. In photoshop there's a video filter offering the option to de-interlace - should that be used?
Thanks & regards.
Gra
MSP8 (SP1), VS8, C3DPS, MF6+, DAZ Studio, Poser 6, Nero 6, Audacity, Photoshop 7.0
You can see a couple of my movies at [url]http://www.youtube.com/glaustin[/url]
Gra
MSP8 (SP1), VS8, C3DPS, MF6+, DAZ Studio, Poser 6, Nero 6, Audacity, Photoshop 7.0
You can see a couple of my movies at [url]http://www.youtube.com/glaustin[/url]
A deinterlace filter is designed to remove the interlacing from a single frame capture of video. No image capture from a digicam or scanner is interlaced, so you cannot deinterlace it, can you?
[b][i][color=red]Devil[/color][/i][/b]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
Only un autre morceau de logique: remember that your final result is probably 720x576/480. Don't expect the same quality as, say. 800x600 or higher.
[b][i][color=red]Devil[/color][/i][/b]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
Thanks. I do resize down to 720x576 and my dpi is always set at 300. Should I leave that DPI setting at 300 or size down or up as well??
Thanks & regards.
Gra
MSP8 (SP1), VS8, C3DPS, MF6+, DAZ Studio, Poser 6, Nero 6, Audacity, Photoshop 7.0
You can see a couple of my movies at [url]http://www.youtube.com/glaustin[/url]
Gra
MSP8 (SP1), VS8, C3DPS, MF6+, DAZ Studio, Poser 6, Nero 6, Audacity, Photoshop 7.0
You can see a couple of my movies at [url]http://www.youtube.com/glaustin[/url]
dpi has no meaning whatsoever in a video context. If an image is 720 pixels wide, it means it fills the screen, whether it is 4 inches diagonal or 42 inches diagonal.
[b][i][color=red]Devil[/color][/i][/b]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
[size=84]P4 Core 2 Duo 2.6 GHz/Elite NVidia NF650iSLIT-A/2 Gb dual channel FSB 1333 MHz/Gainward NVidia 7300/2 x 80 Gb, 1 x 300 Gb, 1 x 200 Gb/DVCAM DRV-1000P drive/ Pan NV-DX1&-DX100/MSP8/WS2/PI11/C3D etc.[/size]
I haven't noticed this loss of quality, and I tend to keep the original resolution because I want to do zooms into the detail at times. The image size does need to be kept less than 5k pixels square because I get out-of-memory on rendering otherwise. I ensure the aspect ratio by using a moving path and checking the appropriate box.
I'm tending to render for progressive these days anyway, and have noticed some nasty artifacts in images rendered for interlaced.
Did you ensure you used bicubic filtering when resizing in photoimpact?
Regards John Baker
I'm tending to render for progressive these days anyway, and have noticed some nasty artifacts in images rendered for interlaced.
Did you ensure you used bicubic filtering when resizing in photoimpact?
Regards John Baker
