[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/diy/ - Do It Yourself

Search:


View post   

>> No.1770480 [View]
File: 331 KB, 444x446, Screenshot_12.png [View same] [iqdb] [saucenao] [google]
1770480

>>1770406
>8-bit parallel interface is plenty fast
So i ran some tests using a cheap 320x240 64k color LCD. I re-wrote some critical bitbanging parts in asm and ran on STMF4 at 168Mhz. The entire GRAM is now updated in about 10ms which would give me about 90-100fps which is faster than I'd ever need. Problem is the flicker since data are written directly to GRAM and I feel like there is no way to avoid it.
> properly memory-mapped
what do you mean by memory mapping? I haven't used FSMC. I ported/hacked some LCD drivers I found on github that bit bang all the strobe signals to emulate the 8080 mode (8-bits plus control signals). I don't know if FSMC would give me any improvement in speed and/or flexibility.

However as a next step I'd like to try something bigger than 320x240 and thats where the 8-bit parallel mode might not be fast enough. I am not looking to play videos, at least for now, just wanted to get a refresh rate close to 60 fps. Which is why I was wondering if it would make sense to buy some of those cheap LCD controllers (like in the pic) but wire it for RGB instead of 8-bit. I don't know if thats a popular approach and how to best drive it. I don't want to try to bitbang RGB+sync signals.

>LVDS
thanks this is interesting, need to read more on that. may graduate to LVDS panels as the next step.

>>1770406
>there are internal pulldowns on that pin but I don't trust them in a noisy environment
not just noise, may lead to glitches due to high/low level thresholds affected by a slow VDD ramp-up.

Navigation
View posts[+24][+48][+96]