SteelSeries 2.5.5 and the Wiki

[Blicj11]

I have edited the SteelSeries Gauges wiki entry to include a more complete explanation of implementing SteelSeries on a website. I have also added descriptions of many of the configuration changes that can be made to size, colour, font, background, etc. This edit is based on the latest release of SteelSeries, which is 2.5.5. If you use these gauges, please take a look at the wiki and let me know of all the typos and any other edits that would make it clearer. The wiki entry is at http://wiki.trixology.com/index.php/Steel_Series_Gauges

Also, if you are still using version 2.4 of the gauges, you are missing out.

[wurzelmac]

The only thing I would change is: gauges.ss.ccs into gauges-ss.css and gauges.ss.basic.htm into gauges-ss-basic.htm - the rest is well done. Thanks a lot for this work, every now and then (especially when upgrading to a new version) I will have a look at this part of the Wiki! As I mentioned in another thread the explanation of the single gauges on your website is well done, too!

Cheers,
Reinhard

[Blicj11]

Good catches Reinhard. Thanks. I went back through the Wiki and found several additional typos. I think it is mostly cleaner up now. Thanks for your comments about the gauge explanations on my website. That website is the official website of the mountain community where I live. Since I put the weather page on the site, the weather page has 4 times as many views as any other page so I guess I wasn't the only person on the mountain interested in the local weather.

[Blicj11]

Well, you're right, it doesn't display like that in Chrome 

Best thing to do if this sort of thing happens is to capture a copy of the realtimegaugesWC.txt file - there is probably something in there causing it - it looks like the problem has cleared now?

Mark:

See the two attached screenshots. This happens just after midnight.

[mcrossley]

Nothing wrong in that data file that I can see, maybe I'll have to set my alarm for tomorrow morning and take a look as it rolls over

Does it happen if the page is running at midnight, or if you load the page at midnight?

Is it only wind chill you have noticed it on?

I don't suppose you noticed if the mouse-over was displaying sensible values?

[Blicj11]

Before you do anything crazy, like setting your alarm, let me do a bit more poking around.

My page was already running at midnight. The first refresh of data after midnight is when the auto scale goes bonkers. I think it only happens on wind chill. It definitely only happens on that gauge. It will automatically correct itself after several minutes. I can force it to correct itself either by reloading the page, or by selecting one of the other options on that gauge and then immediately re-selecting wind chill.

I will play with it for a few nights. I'll select Apparent before midnight and see what happens. I will also check out what happens with the mouse-over. I'll report back here with what I find.

[mcrossley]

OK thanks, I suspect that it may just be the first data download after midnight - the data file you posted was stamped as 7 minutes past midnight, how often does it get updated?

[Blicj11]

That was my fault. I had the first data file after midnight but I accidentally deleted it. Since the gauge was still showing goofy numbers I just used that one. I update every 30 seconds. I'll get it tonight. Unless I fall asleep. 

[WCDev]

That -10000 looks suspiciously like a 'data not available' sentinel value - is it just on wind chill?

[Blicj11]

Tonight I had the page running with the gauge selection set to Apparent. It did the same thing as it did on Wind Chill: The first time the data is refreshed after midnight, the gauge scale went crazy although the reading itself, as well as the mouse-over graph is correct. I changed the selection from Apparent to Wind Chill. The problem persisted. When I reloaded the page, it corrected itself. It only happens with this gauge. See attached screenshots.

[mcrossley]

Ah ha! We have the culprit.

The 'today' highs/lows for all readings are set to 'N/A'.

The 'other' programs set them to the current readings at the day roll-over. It looks like WC does not set them until a few minutes after midnight.

Now how to deal with that... Also I would have expected the scale to reset to the default values specified in the config section, so something to look at there as well.

[WCDev]

I wouldn't bother trying to work around it Mark - it's something I can change in V2.2.0

[Blicj11]

Thanks for spotting that Mark. Stu, thank you for putting this on the list for V2.2.0.

[mcrossley]

OK, but I have fixed the scale going to -10,000, it is because I use -9999 to signify invalid readings, but the dew point gauge code wasn't trapping that return value and then doing something sensible! I'll get that fix uploaded when I have tested it some more...

[Blicj11]

On to my next question. The popup graph on my temperature gauge is different than all the other popup graphs. It looks to me like it has a finer line (or thinner) on the graph that all the other graphs. I am not using a custom graph there. Is that just my eyes? Is it possible I have tweaked that graph setting in SteelSeries and don't recall how or where?