Hollis TX-1 review

[dmaziuk]

Last - price:

Everywhere on the Intertubes has the TX-1 listed for $550 or $950 with the transmitter. I emailed LeisurePro a couple of weeks ago and asked for a price. They quoted me $450 for the TX-1 or $725 for the TX-1 with the AI transmitter. I ended up getting an even better deal from a local dealer. But, even at $450, that seems like a pretty awesome price for a fully capable, trimix, tech computer. After looking seriously at the Nitek Q, which is $400 from DGX, I definitely feel like the TX-1 is worth an extra $50. If you truly care absolutely nothing about AI, or even the potential resale value of the AI capability, then MAYBE the extra $50 is not worth it. Of course, that is really up to each individual.

It may be worth adding, for the future readers, that if you do care about AI, the cheapest available option is Hollis DG03 at $250 -- but the USB cable for it is another $100 and the bungee mount is $20. If those are included in TX1's $400, then you're looking at only $30 difference, and that gets you ZHL-16C w/ GF and a (somewhat usable) compass.

Thanks for the review.

 

[stuartv]

I did 4 wreck dives up in NJ this past weekend with my H3 as primary and my TX-1 as my backup. The TX-1 was not the backup computer I was hoping it would be. Here is the email I just sent to Hollis:

Hello and happy Halloween,

I purchased a new Hollis TX-1 about 1 month ago. I have now done 9 dives with it. The last 4 dives were yesterday and the day before. 2 dives on Sat and 2 dives on Sun. I have now had several occasions of different odd behaviors I'm hoping you can explain.

1) On Sat, I did two dives. The first was to 107' with a total duration of 42 minutes, using 28/22 trimix. My other computer was set to GF of 50/80. My TX1 was set to 30/70, because I forgot to change the TX-1 and I was following my other computer. After I got out from this dive, the TX-1 went into Violation mode.

I downloaded the dive data from both computers into Subsurface. The logged data shows that I never exceeded a GF of 80. I don't understand why the TX-1 went into Violation mode.

What I do see is this info from the TX-1:

At 35:10, I was at a depth of 17.5' and the data says I had a deco time of 1:00 at 30'.
At 35:15, I was at a depth of 17.7' and the data says I had a deco time of 18:00 at 20'. I.e. in 5 seconds of elapsed dive time, it added 17 minutes of deco!

The dive ended at 41:30 and the TX-1 says I still had deco time of 7:00 at 20'.

Since the TX-1 was set to GF 30/70 and the computer I was actually following was set to 50/80, I understand that from the perspective of the TX-1 I probably skipped a deco stop (or 2). But, I have consulted the TX-1 User Manual and as I never exceeded an actual GF of 90, I don't understand why it went into Violation mode.

2) Since the computer was in Violation mode (Violation flashing on the screen while on the surface) after my first dive on Saturday, I did my second dive with it where I gather that it was running in Violation Gauge mode. It wouldn't let me do a gas switch. I used 80% for deco gas on my second dive. Fortunately, my other computer was still working fine and I did the gas switch and followed it for the 2nd dive.

But, the next morning, (way less than 24 hours after the end of the 2nd dive from the day before), the TX-1 was back to functioning normally in Dive mode. I thought once it went into Violation mode, it would stay like that for at least 24 hours.

3) Sunday morning before my first dive, I pressed one of the buttons to wake up the TX-1 and it appeared to go through a reset. At the end, it prompted me to go through a calibration process. It said to put it face down until it beeped, then face up, then rotate it. I did all that. Maybe it somehow fully reset and that's why it was no longer in Violation mode for my dives on Sunday. The History still shows 9 dives and the My Info display is what I set it to, so it didn't completely reset.

4) I did 2 dives in my local quarry just over a week ago. The quarry is pretty consistent on its bottom temperatures. it ranges from 37F to 42F all year round. For the 2 dives I did there last week, my other computer logged the bottom at 42F. That is completely consistent with past experience there and the other computer I used to use before I got the TX-1. On those same dives, the TX-1 logged the bottom temp as 32F. Needless to say, I'm pretty confident the fresh water quarry I was in was not actually 32 degrees.

On all 4 dives of this past weekend, my other computer logged bottom temps of 62F. The TX-1 logged all 4 dives as 56F.

I think the TX-1 is off and it gets further off the colder the water is.

I don't think it matters, but just for the record, the depth reading for the TX-1 and my other computer have consistently matched each other to less than 1' difference at all depths.

Can you please advise me on what to do about this? I suspect this could all be fixed with updated firmware, but when I got the TX-1 a month ago I updated it to the latest firmware.

Thank you for any help you can provide.

- Stuart

 

[stuartv]

According to the TX-1 User Manual, if you miss a deco stop, it will bitch at you. If you exceed the stop ceiling for more than 1 minute, it will beep at you and then it will recalculate your ascent using a "less conservative setting".

But, according to the manual, it would only go into Violation Gauge Mode if you exceed a Gradient Factor of 90 for more than 5 minutes. So, with a setting of GF30/70, but following my other computer set to 50/80, the Hollis should complain at me for missing mandatory deco stops and even think that I got out of the water having skipped mandatory stops. But, since my other computer says I got to the surface with a GF of 80 or less (and the data in Subsurface agrees), and since the timer and depth gauge on both computers agrees, the TX-1 should not have gone into Violation mode - assuming the implementation actually matches the documentation.

And that is, I think, the real issue. I suspect that it's just not implemented to work precisely the way the manual says. I suspect that if I had remembered to set the GF on the Hollis to match what I had set on my H3, the TX-1 would have been fine. Instead, it thinks I got out of the water having skipped a mandatory deco stop and so I think it went into VGM anyway, even though I didn't exceed a GF of 90.

I'll post another update if/when I hear back from Hollis.

I would also sure like to know why it, apparently, reset itself Sunday morning when I turned it on. I checked and it says the Battery Status is "Good", so I can't attribute it to low battery funkiness.

 

[Doby45]

I was really looking forward to the review @stuartv but it seemed to fail right from the beginning with the fact that you did not have the two computers matched. I understand it still should not have gone into violation mode, but from that point on it kinda screwed the pooch as far as seeing if it is a good backup. I would give it another shot for sure with the GF matching on both, which would be the point of a backup anyway. My hope is a quasi inexpensive backup to a Petrel with the ability to add a transmitter on down the road.

 

[stuartv]

Right. I mean, in my opinion, no computer that is intended to be used by tech divers should ever go into VGM at all. Having it set to the wrong GF should just mean being told the wrong deco stops. I should be able to change the GF setting on the fly - and I shouldn't have to change the GF setting to prevent it from doing something like going into VGM.

Of course, it also should not reset itself for no reason and lose all my tissue loading data from the day before.

I made a small mistake (in not setting my GF on the TX-1). But, in the end, I think this kind of turned out to be a good test. Knowing the computer works fine when you do everything right is one thing. Knowing that it will still work fine if you forget to adjust your settings is even better - or in this case, knowing that it might NOT work like it's supposed to.

In reality, since it's supposed to be my backup, maybe I should just set it to GF 90/90 or something like that and leave it. It should always be more liberal than my main computer and if I never need an emergency ascent, I can refer to it without having to do any button pressing at all.

 

[the_ocean]

The TX1 and VT4.1 look like the same hardware, just different software and maybe a slightly different LCD face. Watch out for bugs. I've had a repeated issues on my VT4.1 with a bug that Oceanic says they have fixed in their last firmware release, but it still happens. I have gone through 3 computers with the exact same issue, so it's definitely firmware related.

Here a quote from the last firmware update notice for VT, detailing the bug that they claim to have fixed, which keeps happening:

Corrected premature entry into Deco that could occur on repetitive Nitrox dives conducted after between dive surface intervals during which the Set FO2 mode was accessed, then Plan mode was accessed then exited by pressing M for 2 seconds or after a 2 minute time out.

You can see this by checking the logged TLBG (tissue load bar graph) on a PC. The VT4.1 logs don't include individual tissue information.

Do the TX1 logs include tissue information? Would be great if you could upload a sample log file.

Also, how closely does the TX1 at 90/90 track the NDLs vs a computer running DSAT?

 

[stuartv]

Thanks, @the_ocean. I had not before, but now I have downloaded my dive data from the TX-1 using the Hollis software instead of Subsurface. The Hollis software shows me a little bit more info and in detailed CSV record format, instead of the Subsurface graph.

One of the columns in the data is the number of segments in the Tissue Loading Bar Graph. That number starts at 0 and gradually starts going up during the dive. The data is logged every 15 seconds.

At 15:45 (mm:ss), the TLBG is 3 and my depth was 101'.
At 16:00, the TLBG is 5 and my depth is still 101'.

The seems like a problem right there. Going from tissue loading of "3" to "5" in 15 seconds, while still at the same depth doesn't seem right.

It is at this same point that the Notes column says I went into deco and shows a 1 minute stop at 30'.

During the dive, I continued on, not paying attention to the TX-1 and started my ascent at around the 21:00 mark. The computer still shows 1:00 at 30' for my stop.

I ascended all the way to 28' where the computer then logged than I had a missed deco stop, but still wanting 1:00 at 30'.

The data has a column for ascent rate where it just shows one of "0-10FPM", "0-20FPM", "11-15FPM, "16-20FPM", "26-30FPM", or ">30FPM". Weird the way some of those ranges overlap, but anyway....

During the entire ascent, my ascent rate is logged as 0-20FPM except for one span of 4 records that shows 0-10 followed by >30 followed by >30, followed by 0-10. Again, this is data logged every 15 seconds. Over the span of those 4 records, I went from 55' to 33' - so 22' in 1 minute. In the 30 second period that has 2 records saying >30FPM, I went from 51' to 37'. So, 14' in 30 seconds. 28' per minute. And that was the fastest portion of my whole ascent.

At 34:45, I was at 17' depth, and the computer still says deco of 1:00 at 30'.
At 35:00, I was at 18' depth, and now the computer says deco of 18:00 at 20'.

I guess that was when it decided to recalculate for a "less conservative" ascent.

I finally hit the surface at around the 41 - 42 minute mark. My final ascent from 17' was all at 15' per minute or less. Mostly 10' per minute of less, actually.

I still have no idea why it went into VGM - other than the theory that it is different than the manual and did it because of the deco stop it thinks I skipped (which was based on GF30/70).

Either that or the spot in the data where the TLBG jumps directly from 3 to 5 is an indication that it has a bug that caused it to actually think my GF was higher than 90 for more than 5 minutes at some point.

 

[stuartv]

The TX1 and VT4.1 look like the same hardware, just different software and maybe a slightly different LCD face. Watch out for bugs. I've had a repeated issues on my VT4.1 with a bug that Oceanic says they have fixed in their last firmware release, but it still happens. I have gone through 3 computers with the exact same issue, so it's definitely firmware related.

Here a quote from the last firmware update notice for VT, detailing the bug that they claim to have fixed, which keeps happening:


You can see this by checking the logged TLBG (tissue load bar graph) on a PC. The VT4.1 logs don't include individual tissue information.

Do the TX1 logs include tissue information? Would be great if you could upload a sample log file.

Also, how closely does the TX1 at 90/90 track the NDLs vs a computer running DSAT?

Interesting. Even though the software must be totally different (Buhlmann with GF calculating for trimix versus DSAT calculating for nitrox), this does sound like the same bug.

I'm attaching my downloaded log file for this dive for you and posterity.

As for GF90/90 vs DSAT, the best source of that info is @scubadada. He has been diving DSAT versus Buhlmann+GF (in a Nitek Q) for a few months now. You can search for posts from him. He has posted several pretty detailed accounts of how they match up. The short answer is (I THINK) that GF95/95 matches DSAT about the best, but it's definitely not a perfect match. Sometimes one will give more NDL, sometimes the other.

For my recreational dives, I use an Atom 3.0 running DSAT and my H3 with GF set to something like 50/80. On a rec dive, I will stay down right to the end of the NDL on the Atom, which will generally have me in deco on the H3. Then I will do my ascent based on whatever the H3 says. And do a minimum of 5 minutes between 15 and 20', no matter what. This lets me get back on a recreational charter boat and say with complete honesty that I did not go into deco on my primary computer. And, of course, if they want to look, both computers are clear and happy.

 

[the_ocean]

Interesting. Even though the software must be totally different (Buhlmann with GF calculating for trimix versus DSAT calculating for nitrox), this does sound like the same bug.

I'm attaching my downloaded log file for this dive for you and posterity.

Yours seems like a different behavior (not sure if it's a bug) since it occurs during a dive.

The bug I mentioned occurs on the surface interval following a dive. Using the plan and FO2 features during the surface interval can result in the computer pausing its offgassing calculations until it goes back into dive mode on the next dive.

Here's a plot of the log you showed:

Just wanted to put that graph up for others to follow along, I don't have an explanation on why TLBG jumps from 3 to 5.

I had another VT4.1 issue recently where me and my dive buddy dove together with identical VT4.1 computers and identical profiles. At the end of the dive, I went into the FLY/DESAT menu on both computers and mine showed 3 more hours of desaturation time than my buddy's, which is not right. I checked the logs from both computers and found something fishy: the depth profiles match up, except my log shows 3 additional samples at the start of the dive (15 seconds apart) showing a descent to 15ft and then surfacing. This initial descent did not happen, my buddy and I descended together slowly. We did not splash in from a boat, this was a 5 minute swim out from shore, so I'm not sure how the pressure sensor registered a sudden descent. Regardless, a quick bounce to 15ft should not result in 3 hours additional desat time, so I guess that the pressure sensor picked up some crazy spikes which weren't recorded in the logs but did affect the algorithm (both computers were running DSAT).

 

[the_ocean]

I wanted a backup for my AI Oceanic VT3, which runs DSAT. I wanted to gain experience with Buhlmann GFs along with my primary computer. My next computer will be an AI running GFs, whenever an acceptable one becomes available.

I have the exact same desires - want to switch my VT4.1 to a backup computer (running DSAT or PZ+) and use its AI feature for after-the-dive tracking of my RMV, and for the primary I would like something running GFs, with a better display and better logs (showing individual tissue states) so that I can piece together what happened if anything goes wrong. Looking strongly towards the Ratio iX3M (with or without their soon-to-be announced AI).