Welcome to LinuxCNC.org

andypugh wrote:Thanks for the tips. I'll re-read through the man page again and see what I can learn.

Hi

The file in /etc/X11 is called xorg.conf and the command to generate a new config is Xorg -configure

I updated the wiki a little while ago, in respect of installing the vesa drivers, which is where most people need to edit xorg.conf

wiki.linuxcnc.org/cgi-bin/wiki.pl?Troubl...g#Using_Vesa_Drivers

Now shows a bit more info about creating xorg.conf where it does not exist (as in Ubuntu 10.04 and above by default)

regards

AlexN wrote:
Nice latency, what was the second jitter latency?

What about those big interval numbers? What were they?

ArcEye wrote:
Hello ArcEye, thanks for the link . I've been busy wrestling with other hardware concerns, including RS Australia having a problem with an order payment where I'd actually paid for it - the details were in my online bank records - with their claiming that I hadn't paid for it. they had also rung me me up about it twice, but their calls had sounded like a spam calls (ultra long delay/no answer, so I unknowingly hung up on them bot times)...

I also didn't seem to get the Forum's email headsups for the last two posts and didn't check the thread. Oh well.

Enough of the whinging and whining and bleating and squeaking .

Interestingly, I had a hunch that Ubuntu might be adding in the touchscreen support by default, so I installed a new version on the yawning chasm that is my 80 GB ssd - and lo and behold, the touchscreen worked "out-of-box" with no intervention from me . It does, however, have the X-direction reversed; I'll nibble away at it in between getting the hardware all hooked up.

Speaking of trhe latter, I have to make a new cable per Andy's suggestion near the beginning of this thread - but with a literal twist: I can either use a ribbon cable or a "round" cable, but in either case I am going to have to solder one end of the new cable. Or take to the D525 MB and solder a different connector in place of the standar D25 female connector and thus smoking my warranty on the board (and potentially the board itself).

The other bit with a twist in it is that the SoundLogic board appears to have its pinouts "back to front" (although not upside down), which means being careful tsolder the pins backwards from teh standard pinouts. At least, that's how it looks: I'm going to run through this several times before firing up the soldering iron, as I don't want to have to solder the 25-pin IDC conector twice!

Cheers,
Alex.

gera229 wrote:Here's the full output from that snapshot of the latency sniffer, converted to text:

Setting that table up so that it lined up OK in the Forum preview was really weird, even using the "code" element (and because I used the "code" element, I had to remove the bold formatting )...

BTW, the latencies are around 7000-8000 ns when just running the sniffer by itself, "isolcpus=1", hyperthreading nuked, etc.

HTH,
Alex.

My question was based on your post where you already disabled hyperthreading and already set Isolcpus to 1.

Reread it, I mainly asked if the first jitter latency was no higher than 12656 (you said that's the value you got with it all setup properly with hyperthreading disabled and Isolcpus set to 1), what was the second jitter latency and what values were both of the interval latencies?

Gera