Main > Software Forum
So What Cause's The Stuttering Sound Problem With DirectDraw??
AndyWarne:
--- Quote from: RandyT on August 14, 2007, 09:32:12 am ---I think I know what a pixel is. I designed computer graphics on graph paper in the 80's when that was the only way to do it. :D
--- End quote ---
I know that you know what a pixel is!
What I was getting at is that a missing step in the top of a Pacman, or a double-width step is noticeable. One column is one pixel, not many pixels as you suggested, when native resolutions are being used.
Andy
RandyT:
--- Quote from: AndyWarne on August 14, 2007, 06:55:05 pm ---I know that you know what a pixel is!
What I was getting at is that a missing step in the top of a Pacman, or a double-width step is noticeable. One column is one pixel, not many pixels as you suggested, when native resolutions are being used.
--- End quote ---
No, a column is the number of vertical pixels in the native screen resolution, or in the case of Pac-Man, 288 extra pixels for each extra column that exists in the image. Likewise, each extra row adds 224 extra pixels in Pac-Man each and every time it occurs. When someone says "one pixel" it sounds very much like an insignificant thing, because one pixel out of the 64512 pixels present in that image is insignificant. But we are talking about a thousand or more of extra (or missing) pixels which is very much a big deal.
Semantics aren't a big deal until someone doesn't understand what all the fuss is about.
RandyT
headkaze:
--- Quote from: RandyT on August 14, 2007, 07:37:33 pm ---
--- Quote from: AndyWarne on August 14, 2007, 06:55:05 pm ---I know that you know what a pixel is!
What I was getting at is that a missing step in the top of a Pacman, or a double-width step is noticeable. One column is one pixel, not many pixels as you suggested, when native resolutions are being used.
--- End quote ---
No, a column is the number of vertical pixels in the native screen resolution, or in the case of Pac-Man, 288 extra pixels for each extra column that exists in the image. Likewise, each extra row adds 224 extra pixels in Pac-Man each and every time it occurs. When someone says "one pixel" it sounds very much like an insignificant thing, because one pixel out of the 64512 pixels present in that image is insignificant. But we are talking about a thousand or more of extra (or missing) pixels which is very much a big deal.
Semantics aren't a big deal until someone doesn't understand what all the fuss is about.
RandyT
--- End quote ---
"One pixel" might sound insignificant to someone who has no idea what is been talked about here. I think it does the job. I don't think "one vertical column of pixels" is necessary unless were trying to explain the concept to kindergarten children. If you look at the scaling problems in the picture "one pixel off" should suffice whether that be a vertical column or horizontal row of pixels. It's an excercise in futility.
And I don't think you were the first person here to design graphics using graph paper either ;) It doesn't help prove your point either.
RandyT:
--- Quote from: headkaze on August 15, 2007, 01:41:52 am ---"One pixel" might sound insignificant to someone who has no idea what is been talked about here. I think it does the job. I don't think "one vertical column of pixels" is necessary unless were trying to explain the concept to kindergarten children. If you look at the scaling problems in the picture "one pixel off" should suffice whether that be a vertical column or horizontal row of pixels. It's an excercise in futility.
And I don't think you were the first person here to design graphics using graph paper either ;) It doesn't help prove your point either.
--- End quote ---
I never said I was and now you are just being snide.
Take a look at the image ahofle posted again. There are no less than 8 extra rows of pixels, resulting in a total of 1792 extra pixels on the screen. You don't have to be a "kindergartener" to know that this statement far better represents the situation than speech like "it's only one extra pixel."
It's reckless to assume everyone who reads these discussions has the same background you do. If you do, you'll have individuals parroting what is written without fully understanding the ramifications, and the rest of the folks to whom these issues really matter will be the ones left in the cold or painted as "anal retentives". If you don't care about artifacting and the damage it causes to the images presented in the games, that's fine. It's your prerogative. But please don't minimize the situation to others, who may not fully understand it, with incorrect language that they may not have the experience to know is incorrect.
RandyT
headkaze:
--- Quote from: AndyWarne on August 13, 2007, 03:05:25 pm ---I have not yet found a way to turn off all scaling in D3D.
As an example, if you consider a game which runs at 304 X 256 and you run this at a screen resolution of 320 X 256, using Ddraw results in an exact map of the game pixels onto the screen, with an un-noticeable border of 8 pixels either side.
If you try this with D3D, it attempts to re-scale the 304 pixels onto 320. This results in vertical lines of pixels being duplicated to fill out the spare space. The result of this is a distorted picture and is very noticeable.
If anyone can find a way to disable all stretch in D3D this would be very useful.
--- End quote ---
Have you considered the artwork folder in Mame which is used for bezels? Perhaps writing some program to generate special bezels to create the borders required for each game so Direct3D will stretch the image into a rectangle to match the resolution (or an even multiple thereof for perfect scaling)?
Check out a normal *.lay file for 1942. Perhaps this could be used for this purpose?
--- Code: ---<mamelayout version="2">
<element name="bezel">
<image file="1942_bezel.png" />
</element>
<view name="Upright Artwork">
<screen index="0">
<bounds x="1036" y="647" width="1860" height="2480" />
</screen>
<bezel element="bezel">
<bounds x="0" y="0" width="3960" height="4000" />
</bezel>
</view>
</mamelayout>
--- End code ---
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version