Widescreen Gaming Forum

Wasn't really sure where to post this one, feel free to move this thread if it's in the wrong place!

I've seen the attention that MView and TH2Go etc. has, and the equal confusion on what resolutions can or can not be used.
Hopefully the following will at least partially help make things more clear.

A new tool has been added to the site's Projects under the Widescreen section. The Widescreen Bandwidth Calculator should help you determine how much bandwidth a resolution needs, thus allowing you to more easily figure out what type of display cable you need for your monitor setup.

There is also a list of the most common display cables and their bandwidths.

You can find it in the Projects: Widescreen section.

As your site states, dual link should support 5040x1050. But there is also a small buffer/reservation in bandwidth, and that is why sometimes the resolution isn't supported while the bandwidth is supported.

But overal, good tool! :onethumb

Well! That's the video bandwidth, the cable has even more.
DVI and HDMI has the sync info seperate while DisplayPort has it with the video.

If these numbers are wrong blame the standards guys inflating the spec numbers :P

The calculation though should be accurate.

The v1.2 DisplayPort look really interesting, and hopefully we'll see updated TH2G and MView etc. taking advantage of that.

I did some interesting calculations. Matrox / Mview could do some interesting stuff with DP 1.2 as it has a much higher Bandwidth.

The highest usable resolution at 60Hz would be 7680x1440 (3x2560x1440) (I.E. the 16:9 version of those newfangled 30"rs)
(The highest possible Resolution at 60Hz would be 7680x1563 which is kinda useless unless your panel has awesome scaling.)
The highest usable resolution at 58Hz would be 7680x1600 (3x2560x1600) (I.E. the newfangled 30"rs)

A new Revision of the TH2go to DP1.2 TH2go would be able to do 5760x1200@60Hz no problem. So guys it looks like if Matrox / Mview want to keep afloat, we will see something like this:

DP1.2 Connection to PC from DP1.2 TH2go
3 DP+ Connections from the DP1.2 TH2go to monitors. It would killer if we had to use 3 x $100 adapters for a box that will already cost out the butt.
Pass-thru DP Audio, to 1 or all 3 output Ports

I personally like Matrox, because they let you choose a larger variety of resolutions than eyefinity. Though Matrox should learn from the current market, that 16:10 monitors are going the way side and should support more 16:9 resolutions as well as 16:10.

_________________
System Core: | Intel Core i5-2500K + ASUS P8Z68-V + 16GB Corsair XMS3 DDR3 1333 MHz | Win7 x64 | MSI R7970 Lightning 3GB [1105/1400] |
Display: | 3 x Dell Ultrasharp 3007WFP-HC @ 7680x1600 | Dell u3011 |

Lemme Zombie this thread as Matrox might have a chance. Lookie at this link http://www.engadget.com/2010/01/07/dell-goes-pro-with-27-inch-ultrasharp-u2711-wqhd-lcd-monitor-ha/

Dells made an ultrasharp that is in fact 2560x1440, I should bring this up to matrox.

_________________
System Core: | Intel Core i5-2500K + ASUS P8Z68-V + 16GB Corsair XMS3 DDR3 1333 MHz | Win7 x64 | MSI R7970 Lightning 3GB [1105/1400] |
Display: | 3 x Dell Ultrasharp 3007WFP-HC @ 7680x1600 | Dell u3011 |

Someone named Rhy K used the contact form on my site and told me my calculation was off by 25%.
This is understandable as no overhead is taken into account.

However after researching some...
I found http://www.csgnetwork.com/videosignalcalc.html
there is also http://www.epanorama.net/faq/vga2rgb/calc.html
and http://xtiming.sourceforge.net/cgi-bin/xtiming.pl

and after checking with the different timing modes the graphics driver support (in my current system uses),
like CVT reduced blank which uses the least bandwidth, through DMT, CVT, and GTF which uses the most,
And the fact that progressive and interlace also impact the bandwidth. (interlace uses slightly less than progressive for example).
And after testing edge cases, I found that the "sony" algo is close but a little off (might be the rounding?).

The calc did not just need a 25% increase, in fact it needed a whooping 50% overhead to be added. (outch)

This is confirmed by the 1920x1200@60hz screenmode which uses around 207MHz bandwidth,
which is what this monitor for example http://benq.se/products/LCD/index.cfm/page/specifications/product/1156 supports as max.
205MHz also equals 4.98Gbit, and the HDMI 1.0 and 1.1 and 1.2 standard has a max 4.95Gbit for TDMS data.
This matches what some has said, that 1920x1200@60 didn't work but 1920x1200@50 (or @59) did work etc.

I've updated the bandwidth calc to reflect this, and I doubt it can be made any more correct using only width,height,hz as input variables.

(Note! Since the WSGF FOV & Aspect Calulator's bandwidth calculation/info is partly inspired by my calc it too will need to boost it's bandwidth result by approx. 50%)

Actually, i had to mess around even more. No two sources give the same numbers or calculations.
They mix raw bandwidth with the various timing modes, and fail to note which is which.
Wikipedia is really horrible on this.

I ended up just doing the MHz plain, and the Gbit now has a 3.56 modifier in addition to the value scaling. *sigh*
I may revisit this mess later, but at least it's more accurate than before, and results should be among the worst/most bandwidth eating timing methods.

To mess up your head even further, http://www.mtbs3d.com/phpBB/viewtopic.php?f=3&t=5512&start=0
I can't believe that table is correct, and also those values are based on a VESA document, but that link is no longer working. *groan*