@CullenShane - Most of the Motorola radios with rechargeable batteries also run on either AA or AAA batteries. When the rechargeables in New York Region's crapped out after 3 years, I just picked up a brick of alkaline batteries for every event and tossed the batteries into the recycle bin at the end of the day. I ran our Farmtek timer/photoheads for 8 events last season on a single set of Harbor Freight alkalines, and they have sales every few months where bricks of those are $4.99.
With that said, you could probably also get away with just picking up a set of AA/AAA rechargeable batteries instead of paying the "Motorola tax" to have them sell you what is most likely the same batteries in a fancy plastic case. I never tried it, because the convenience of using disposable batteries out-weighed the hassle of having to find 5 free outlets to plug in the radio chargers the day before every event.....
I would love to get Announcer screen to where anyone could announce cool stats. Not just about the current run but also about how they are doing on the day, past couple events, season, etc.... There is some really cool things we could do here. Suggestions are highly encouraged!!
I'd like to have an announcer function that did the actual announcing! We cannot spare a worker to sit at a microphone at many of our events, and an artificial voice reading times/penalties input to the PA system would be a nice feature.....
Unless something has changed in the off-season, AXti.me is only configurable to use the same COM port that it uses for the timer to also interface to external displays. I have had good luck using one of these with our Farmtek timer:
Since someone had misplaced the DB-9 serial cable that came with our RA display (we were running it plugged into an AXWRATS timer via RJ-45), I also picked up one of these and made an adapter so that a long ethernet patch cable could be used to connect the display:
No dip switch settings on the display were altered from what worked in AXware, and we went back-and-forth between AXti.me and AXware all last season since we still needed to use AXware for Rallycross events.....
The reason being is that obviously PAX calculations are done with Street Touring cars running on street tires while Street Prepared and up cars on r-comps. Grouping these classes together and applying PAX would therefore be 'fair'.
Unfortunately, there are factors in play (weather and temperature) that make any mixing of street tires and R-comps not as cut-and-dry where this is a fair grouping. I'd bet that street tires would out-perform R-comps in a heavy rain with standing water, and can tell you from many late-season autocrosses with temperatures below 45F (7C) that R-comps can't get enough heat in them from an autox run to even stick on par with street tires. We had a debate over this when devising our own local groupings, and defining the groups so that everyone was on similar tires was a primary goal.....
- I setup windows internet connection sharing on the main laptop so that it would act as the DHCP server. The router (Asus RT-AC68) was acting as a WiFi access point only. This was not ideal as I would have preferred having the Router act as DHCP provider, but in that role, all internet traffic would go there, rather than through the iphone. If anyone knows how to fix this, please let me know.
First off, congratulations on running your first event with AXti.me!
The question that I have to ask is why your role computers need Internet access? As long as your on-site router is handing out IP addresses via DHCP and optionally running a local DNS server pointing towards your AXti.me server, the clients should not need to send traffic through your iPhone uplink (and probably shouldn't have access so they don't run your phone bill up downloading updates and stuff).
Of course, you're probably wondering how the live.axti.me results will be uploaded as cars run? Plug your timing server into the router with an ethernet cable, and set a static IP address + subnet mask on the timing servers wired interface with no default gateway defined. Then, have the timing server connect to the iPhone via wifi or USB, which will set that connection as the default route and allow the timing server to send the times to live.axti.me. Packets to/from the local role computers will flow over the local network (without needing a default gateway), and traffic bound for the Internet will go over the cellular interface via the default gateway. Since only the timing server has cellular access through your event network, your cellular bandwidth isn't being wasted.....
Back when I was New York Region SCCA's Solo Chair, I convinced the membership that we should stop handing out attendance trophies (one driver classes) and create some competition. Our ladies and novice classes already ran PAX, so we just made a few more PAX categories:
Stock (now SCCA Street): SS, AS, BS...HS
Street Prepared: ASP, BSP...FSP, SM*
Builder: AP...GP, AM...FM, FSAE, CAM, etc.
The categories have been tweaked a few times since I left, but the general idea is that the fastest drivers (by PAX) get both the event trophies and season points in each category...and we don't have to go through the headaches of bumping to fill classes. As an added benefit, we have a small group of drivers who bum rides instead of driving their own cars at our events, and they like this arrangement because they can run in any class within the category and still collect season points.....
I do remember "TTL..." in the window when it crashed. Now I regreat not capturing it. And you're right, it's step2 that crashes.
I had the same thing pop up on our system two weeks ago (Windows 8.1 32-bit). It ran fine for the first event on 2.3.5, but had the above problem more than once on 2.3.6. Even worse, it took a few tries to get step 2 to restart without exiting immediately again.
I saved the error message in one of the command windows on the server console, and was going to cut/paste it into a file as soon as the event ended. Unfortunately, someone bumped the power switch on the UPS right after the last car finished running, and it dropped power to the server before I could grab it. All that I remember was TTL, so we're both seeing the same problem (and I don't think that it's wireless).....
I just tried to run the web-based system upgrade utility. It failed while building the 2.3.6 web site, logging this info on the console screen:
Backing up data...
creating zipped file : ../_backups/axtime-backup_pre_upgrade_from_2-3-5_2015-09-
Creating temporary backup file ../backups/axtime-backup_pre_upgrade_from_2-3-5
backup events done in 141ms
backup runs done in 578ms
write backup data done in 2157ms
done writing backup data file in 2173ms
Initiating upgrade: from 2.3.5
downloading new build to C:\axtime\website-2-3-5/../tmp-upgrade-bkQ703nW4x-EkSsk
done downloading in 13735ms
extracting files to build-website-2-3-6
done extracting in 90081ms
Copying current settings to new
writing new settings..
Copying 2 uploads to C:\axtime\website-2-3-5/../build-website-2-3-6/public/uploads
Upgrade had errors. Error: ENOENT, no such file or directory 'C:\axtime\build-website-2-3-6\public\uploads\8b32405458ecd57f52978dd08e837d64.def'
socket disconnect. online:0
There is no "uploads" folder in that directory path. Any suggestions how to proceed with the update?????
There is a lot of out of the box thinking with AXTI.me and it really improves over a lot of "standards" and the old ways of doing things, however, printed results have been following a fairly standard format for quite a while...
Actually, the results format that you consider "standard" is really only used by SCCA National and other large events. It is an evolution of the old TS98 results report, now being used in Pronto and autoCRON.
While these formats work well at a national event where everyone gets three runs, they waste 1/2 of the paper real estate with "worthless" information at a regional level and don't scale well to larger numbers of runs. In our club, we don't waste our time even entering car make/models. If it comes in from MSR, it is on the page. Otherwise, it's a whole lotta nothing (like these AXware results from last season):
Personally, I like the simplicity of the AXti.me report (except the classes being alphabetical), because it works well both on paper and on the web...with lots of runs:
Having the car number in its own column has benefits, but it also wastes real estate with fixed columns that is saved by tacking the car # onto the driver's name. I can see arguments for it either way. I could also see arguments for optionally dropping the "diff" columns to save real estate.
One thing that I did not notice until you pointed it out it is that there is no mention of a driver's class in a PAXed category, such as the novice group in the last link above. This is something that I feel should be on the results, since I have no idea which class any of those drivers were competing in without looking at the registration data.....
Also, are there other things we should consider (e.g., tripods to hold WAPs)?
When we do rough site planning for a new wifi deployment, we visualize each person as a standing 6-foot tall column of water that radio signals have a tougher time penetrating. Indoors, we always place an AP either on the ceiling or at least 2.5 meters above the floor on a wall. If you can find a tripod that high, it would certainly help your coverage out.
With the Farmtek timer that we purchased last fall, I am using an external omni-directional antenna with a 3-meter extension cable on the timer box. The photohead radios are still using their factory rubber-duckie antennas. I clamp the timer's antenna to the side of the trailer with a pair of vise-grips. If we used a popup tent, I would clamp it at the top of one tent leg, just to get above the people standing anywhere between T&S and the start/finish photoheads. We really don't push the range beyond 100 meters, and have had zero missed trips wit this setup. In another post after our first event with AXti.me, I even pointed out that our event ran with zero re-runs for any reason...with the Farmtek timer in my truck using the factory rubber duckie and not even the external antenna.
Although this doesn't apply to your situation because you have the RaceAmerica T-links, I will add for Farmtek users and anyone considering an extension antenna a caution about cable length. My 6 dBi omni-directional antenna loses 51% of it's signal in just a short 3-meter extension cable. If I had to go much higher, it would actually be performing worse than the factory rubber duckie. The same applies to wifi, with this same 3-meter cable killing 75% of the signal at 2.4 Ghz and 93% at 5.8 Ghz. In other words, move the AP/receiver, don't extend the antenna cable.....
To answer the network hardware question, this is what I picked up for our base router:
It comes with DD-WRT pre-installed, and most important to me it is dual-band.
We have only run it at one event so far, and that was under less than ideal conditions. (It started raining just as the first car pulled up to the start line, and we had to run T&S from my pickup truck with the access point inside the cab.) They claim it is a high-power AP, but I can't vouch for the range under the conditions we were running it. The majority of problems that I've seen professionally with wifi connectivity issues is that the clients are "weak talkers" due to being in portable devices running on battery with hidden antennas -- not the AP's transmit signal strength.
We are not using any extension APs, nor do I feel that any of our current sites warrant one. The wifi router covers the spectator area and grid, and anyone who wants to see their results from the far reaches of the paddock can walk a few steps closer to the course rather than the club spending money and setup time for additional APs. I might consider adding a battery-powered wireless wifi range extender to the paddock if we ever use the tech inspector role in AXti.me in the future, but financially it probably makes more sense to move tech closer to the existing AP.
I work professionally in IT, and have not purchased anything without 5Ghz wifi since the month before 802.11a was ratified in 2000. A small part of my day job is managing a wifi network spanning 30+ buildings, with about 375 access points and 6,000 devices connected at some point each day. This is what I consider enterprise-grade, and most clients connect via 5Ghz because 2.4Ghz is a train wreck outside of a single AP or home use. With that being said, Ubiquiti is on my no-buy list for two heinous mistakes. (1) Labeling an AP without 5Ghz radios as enterprise-grade. (2) Using a proprietary PoE setup and not supporting the 802.3AF standard in the 21st century. If you are going to use the ones that cost around $100, you can omit a PoE switch and use their included proprietary PoE injectors. If you wish to use them with a standard 802.3AF PoE switch, you'll need to drop an additional $20/AP for this converter:
Ubiquiti's $200+ "Pro" APs are dual-band and may have standard PoE, but there are plenty of other options in this price point.
With the computer requirements, the hardest thing to overcome is "AXware-think" when planning your new setup. You are not chained to a single computer! I ran our first event with AXti.me from a NUC-type mini PC with no monitor/keyboard attached, tossed into the back seat of my pickup. Other than needing to be close enough to connect to the timer with a USB cable and maybe a wired ethernet port, you can put the thing anywhere. The beauty of AXti.me is that it is used from web browsers, with no proprietary software to set up, clunky network shares to attach, etc. My wife ran registration from her personal Windows notebook computer, and later did car queuing from both her computer and her iPhone. I did timing and cone-counting from my personal MacBook Air, as well as an old HP Touchpad from my junk drawer running webOS. If you ask your core workers to BYOD (bring your own device) to each event, you won't even need to buy computers other than something basic to run as the AXti.me server. As an added plus, nobody will be cursing the T&S notebook's keyboard all day because the cursor keys and shift/enter are in the wrong spots from what they're used to. :-)
Cost-wise, I have funded this project out of my own pocket so far. $700 for the software, $70 for the wifi router, $201 for the NUC-type computer and a wireless keyboard/trackpad (just in case), and $45 for a cheap UPS since the NUC-type computer does not have a battery to survive generator stalls. I'm in for $1,016, thanks to dumpster-diving a few things like an old flat-panel LCD display in hopes that we will use the NUC as our main T&S screen when it's not raining.
Those Honda inverter generators are the cat's meow. New York Region had just purchased one right before I took over the Solo program, and it was the only equipment that I could count on at every event. It was quiet enough in eco mode to run under the table if needed, and most people didn't even know that it was at the site until the laser printer's fuser started to heat up before a print of the results which caused it to come out of eco mode. I never ran a UPS or even a surge suppressor with it for several years, and no equipment was damaged from power problems.
My current goal (no pun intended) is to get everything powered from a 12V deep-cycle battery, and not even have a generator. The NUC-type computer that I bought was chosen specifically because it ran from external 12V, as well as the Buffalo wifi router above. Our Farmtek timer run from it's own batteries, the big RaceAmerica display is 12V, and even our PA system has a 12V power option. The only thing that needs a generator is the laser printer, and I'm hoping that after a few more events with AXti.me our drivers will be comfortable enough with having wifi access to the live results where they won't even miss the printouts.....
Could it be because at started my event number at 5? I still have 4 events to import from earlier this season. Is this something to be concerned about?
That's exactly the reason. I changed my first event from #2 to #1 to make the re-calc error go away until I can import the missing first event.....
@Chris - Yup, that fixed it. (Footnote that the problem also happened on iOS 8.0.2 on my test iPhone5, so it wasn't an 8.4 problem.)
We do not use run groups (yet) in the software. After registration, I was playing what-if with the assign run group by class screen, trying to figure out a heat split so that we would have enough workers to cover the course in each group. (With only ~40 drivers at an event, it needs to be more precise to keep things equal.) I did apply it to existing registrations when done.
Even more interesting is that after the event I went and changed the run group labels and re-applied them to everyone. I had originally named them 1/2/3, but some people were getting confused thinking that the run group was a class standing so I changed the group labels to A/B/N after the event and applied it to everyone since they didn't change for drivers just from a re-calculate event. Long story short, the missing driver's run group was apparently missed at least twice, and his class (HS) was most definitely in the red run group on the assign by class screen. I don't think that he was one of the duplicate drivers that I had cleaned up, but maybe that's the case?????
@Chris - TestDataCQcrash contains the event I'm using. I took last Sunday's event (from 2.3.5) and changed the date to today.
I also just discovered that the CQ app starts fine with node.js not running, albeit with the red skull and crossbones. As soon as I start up node.js again, the app closes.
I'm currently reviving my old iPhone5 development/test platform from the dead, to see if this is a problem specific to the 6+.....
From the node.js terminal window:
socket connected: 1
socket sessionId: undefined
socket authorized as anonymous
558ded277c18d8e40add9ee1, Car Queuer, pscc
socket: authenticate success for Car Queuer
socket disconnect. online:0
I think that I should change my forum name to @CaptainCoredump. :-)
I downloaded the new iOS CQ app on my iPhone 6+ w/iOS 8.4 (12H143), and updated the AXti.me install on my Macbook to 2.3.6. The CQ app opened to a login prompt, in which I provided the IP address of my Macbook and the CQ role password. After tapping go, the CQ screen started to open and then the app closed before it could finish opening. Whenever I tap on the app icon, the screen tries to open and immediately fails again.
The first thing that I did was ensure that the old V1.2 version was removed (it was). I tried changing the CQ password in the event, but the app still will not load. The last thing that I tried was to delete the new CQ app and re-install from the App Store. Same result, it asked for my IP and CQ password, and then closed up shop as soon as I tapped go.
One thing that I just noticed is that the splash screen is still running after the failure if I double-click the fingerprint reader button. I tried swiping it up to close and re-launching with no change of behavior. When I tap on the splash screen to bring up again, it still immediately closes the window.
FWIW, I have Test Flight on my phone if you want to push out a debug version and gather info on what's happening.....