Suunto Vyper **SERIOUS BUG** in CNS O2 computation
- Thread starter Genesis
- Start date
Please register or login
Welcome to ScubaBoard, the world's largest scuba diving community. Registration is not required to read the forums, but we encourage you to join. Joining has its benefits and enables you to participate in the discussions.
Benefits of registering include
- Ability to post and comment on topics and discussions.
- A Free photo gallery to share your dive photos with the world.
- You can make this box go away
One after another, here we go.
1. It is possible to get hit at any PO2 over about 1.2. Possible. Just like its POSSIBLE to get a DCI hit while WELL within the tables. You can suffer an O2 hit at any time at ANY PO2 over about 1.2. CAN. Its POSSIBLE. The NOAA exposure limits are there specifically for the purpose of controlling risk, just as the NDL tables are. A dive of 20 minutes at a PO2 of 1.4 is VASTLY conservative, in that it consumes less than one seventh of the NOAA table CNS clock. That is EXTREMELY conservative by ANYONE's estimate. With that said, Jamei, I challenge you to find ONE documented O2 tox hit at or below a PO2 of 1.4 in a recreational diver where the post-incident investigation did not show the presence of an antagonist drug in their system. I bet you can't in the DAN incident reports - I haven't been able to. And no, grabbing the wrong bottle on a gas-switch doesn't count.
2. There is a published NOAA exposure limit for PO2s up to 1.6. Above 1.6 you're on your own, and it is VERY dangerous. The 1.4 recommendation for the "working" portion of the dive was adopted by NOAA and the other agencies both as a CYA protocol and to cover divers who are cold, dehydrated, out-of-shape, or who (apparently!) took Sudafed and other drugs that might accelerate the risk of an O2 hit - and, if a death last year at Ginnie is to be believed, without success (diver toxed long after leaving the bottom, with the max PO2 apparently under 1.3) However, with that note, I will add that it is standard protocol to recompress people in the chamber at a PO2 of 2.0 (yes, that's not a misprint!) and while O2 hits DO occur during such sessions, they are (1) not serious when dry, and (2) do not occur very often, even this far beyond the "safe" limits.
3. This bug exists to one degree or another at ALL setpoints on the Vyper when you approach the MOD. While less severe, I can provoke this bug with a setpoint of 1.3 as well, although the degree of the acceleration is not as serious it still happens.
The problem appears to be a rounding error as I noted, where the computer is calculating the O2 loading under the assumption that you have exceeded the PO2 setpoint (thereby triggering their "acceleration") without warning you that you have exceeded the setpoint (thereby leaving you in the dark that you have entered the "accelerated" mode.)
The EXACT combinations that trigger this and the depth ranges that trigger the worst instances of it are difficult to determine. For example, the aforementioned 32% example, 1.4 setpoint, at 107' is a really ugly example. At 36%, 1.4 setpoint, 91' (their MOD) it doesn't happen at all! At 36%, 1.3 setpoint, at the MOD it happens, but much more slowly (cutting the NOAA limits by about half.)
I've run a dozen simulations at this point and while I can trigger the behavior in most of the scenarios that I've run (but not all), the severity changes from not-terribly-significant (the 1.3 setpoint on 36%) to VERY significant (the 1.4 setpoint on 32%.)
The fact that one cannot predict IF it will bite you on a particular profile, how badly it will bite you, or if it will affect you at all is one of the worst "features" of this problem.
The CNS acceleration in and of itself isn't too cool beyond the setpoint, but at least its documented. What makes this a fairly serious bug is that the acceleration is being triggered without warning you that the setpoint has been exceeded, whcih can lead the diver to the conclusion that he has exceeded CNS limits when in fact no such violation has occurred.
1. It is possible to get hit at any PO2 over about 1.2. Possible. Just like its POSSIBLE to get a DCI hit while WELL within the tables. You can suffer an O2 hit at any time at ANY PO2 over about 1.2. CAN. Its POSSIBLE. The NOAA exposure limits are there specifically for the purpose of controlling risk, just as the NDL tables are. A dive of 20 minutes at a PO2 of 1.4 is VASTLY conservative, in that it consumes less than one seventh of the NOAA table CNS clock. That is EXTREMELY conservative by ANYONE's estimate. With that said, Jamei, I challenge you to find ONE documented O2 tox hit at or below a PO2 of 1.4 in a recreational diver where the post-incident investigation did not show the presence of an antagonist drug in their system. I bet you can't in the DAN incident reports - I haven't been able to. And no, grabbing the wrong bottle on a gas-switch doesn't count.
2. There is a published NOAA exposure limit for PO2s up to 1.6. Above 1.6 you're on your own, and it is VERY dangerous. The 1.4 recommendation for the "working" portion of the dive was adopted by NOAA and the other agencies both as a CYA protocol and to cover divers who are cold, dehydrated, out-of-shape, or who (apparently!) took Sudafed and other drugs that might accelerate the risk of an O2 hit - and, if a death last year at Ginnie is to be believed, without success (diver toxed long after leaving the bottom, with the max PO2 apparently under 1.3) However, with that note, I will add that it is standard protocol to recompress people in the chamber at a PO2 of 2.0 (yes, that's not a misprint!) and while O2 hits DO occur during such sessions, they are (1) not serious when dry, and (2) do not occur very often, even this far beyond the "safe" limits.
3. This bug exists to one degree or another at ALL setpoints on the Vyper when you approach the MOD. While less severe, I can provoke this bug with a setpoint of 1.3 as well, although the degree of the acceleration is not as serious it still happens.
The problem appears to be a rounding error as I noted, where the computer is calculating the O2 loading under the assumption that you have exceeded the PO2 setpoint (thereby triggering their "acceleration") without warning you that you have exceeded the setpoint (thereby leaving you in the dark that you have entered the "accelerated" mode.)
The EXACT combinations that trigger this and the depth ranges that trigger the worst instances of it are difficult to determine. For example, the aforementioned 32% example, 1.4 setpoint, at 107' is a really ugly example. At 36%, 1.4 setpoint, 91' (their MOD) it doesn't happen at all! At 36%, 1.3 setpoint, at the MOD it happens, but much more slowly (cutting the NOAA limits by about half.)
I've run a dozen simulations at this point and while I can trigger the behavior in most of the scenarios that I've run (but not all), the severity changes from not-terribly-significant (the 1.3 setpoint on 36%) to VERY significant (the 1.4 setpoint on 32%.)
The fact that one cannot predict IF it will bite you on a particular profile, how badly it will bite you, or if it will affect you at all is one of the worst "features" of this problem.
The CNS acceleration in and of itself isn't too cool beyond the setpoint, but at least its documented. What makes this a fairly serious bug is that the acceleration is being triggered without warning you that the setpoint has been exceeded, whcih can lead the diver to the conclusion that he has exceeded CNS limits when in fact no such violation has occurred.
Try reading for content.
I have reported this to both Suunto and the LDS that sold me the computer, and have so noted here. Neither have (yet) responded.
If they do not in a reasonable time, or try to duck the issue, I may well take it to the CPSC. My LDS cannot just "hand me another one", since the "other one" will have the same problem!
BTW, you now have three divers with DIFFERENT Vypers who claim to be able to reproduce this. Myself, O-ring (here) and Rich (over on SDN).
Exactly how many "bad" samples do you need to see to be convinced that this is not a "bad computer", but rather a programming problem that affects all of that model?
I am not "spreading misinformation about the tables." 17 minutes at a PO2 of less than 1.4 (but more than 1.3) is less than one eighth of the NOAA-recommended maximum exposure limit, and is VASTLY conservative. For the Vyper to tox you out at that point in time is BROKEN.
Plan the dive, dive the plan... use or have any clue about the content of your tables and you'll know you're safe. Another keystone to good diving. You can't protect bad or unsafe divers from themselves. Genesis, again, have you went by the LDS you bought the Vyper from? How old is this vyper/how long have you owned it? I keep asking about the LDS and you keep ignoring me... That seems like the first place to go to me.
You reported it? You didn't go by and show them? I thought that you said earlier that they "fixed" the problem on newer vypers...
You reported it? You didn't go by and show them? I thought that you said earlier that they "fixed" the problem on newer vypers...
liormic
Guest
that we're somewhat guessing and it looks like Genesis is right and he found a bug at the Vyper's algoritm.
I guess we'll have to wait to suunto's replay.
I guess we'll have to wait to suunto's replay.
Look, I'm not saying that there isn't a bug at all. But you will be hard pressed to find any product/program that doesn't have a bug. Finding that bug may be quite hard. Genesis just found the needle in a hay stack. If the problem is there in every single vyper out there, why haven't we heard of this problem before and why haven't we heard about folks dieing because of it?
NEWS
IMPORTANT SAFETY NOTICE REGARDING SUUNTO VYTEC DIVE COMPUTERS
Despite an extensive pretesting program by Suunto and by independent divers all over the world, Suunto test program has found a software bug in the first series of Suunto Vytecs (products with serial numbers 205000-223700). It has been discovered that using Vytec´s SIMPLAN function may cause problems.
Although diving without using the SIMPLAN function does not cause problems and all features of the instrument work then normally, Suunto wishes that all users contact their local dealer, which contact information you can find here at www.suunto.com or Suunto HQ tel. +358-9-875870 to have their unit replaced with a new unit. Please note that the transmitter unit has no problems and is not to be replaced.
Does your vytec fall into this category? The statement about simdive being different from the actual dive calculations says a lot to me. The simdive may be the only part of the vyper that has a bug (which isn't that important). I know you said you've caused the bug to pop up while actually diving though, so can you submit the downloaded dive information for those divesto us? It would be interesting to see. I know you'll say that the divemanager will alter the calculations, but it will only do that if you hit recalc so you should be able to produce the erroneous dive/dives intact. Thats what you need to get in Suunto's hands. Anyone else done such a dive with a Vyper or willing to?
IMPORTANT SAFETY NOTICE REGARDING SUUNTO VYTEC DIVE COMPUTERS
Despite an extensive pretesting program by Suunto and by independent divers all over the world, Suunto test program has found a software bug in the first series of Suunto Vytecs (products with serial numbers 205000-223700). It has been discovered that using Vytec´s SIMPLAN function may cause problems.
Although diving without using the SIMPLAN function does not cause problems and all features of the instrument work then normally, Suunto wishes that all users contact their local dealer, which contact information you can find here at www.suunto.com or Suunto HQ tel. +358-9-875870 to have their unit replaced with a new unit. Please note that the transmitter unit has no problems and is not to be replaced.
Does your vytec fall into this category? The statement about simdive being different from the actual dive calculations says a lot to me. The simdive may be the only part of the vyper that has a bug (which isn't that important). I know you said you've caused the bug to pop up while actually diving though, so can you submit the downloaded dive information for those divesto us? It would be interesting to see. I know you'll say that the divemanager will alter the calculations, but it will only do that if you hit recalc so you should be able to produce the erroneous dive/dives intact. Thats what you need to get in Suunto's hands. Anyone else done such a dive with a Vyper or willing to?
Brules
Contributor
Can anyone say Uwatec? Look how long it took to find the very REAL problem in their older Ai X Nitrox computers. These are pre 1995 models, and in 2003 they are JUST NOW going to court. Luckily the death/inkuries were not widespread (that they know of), but any defect should be fixed immediately, no matter how small.
The reliance upon the diver diving the plan and not screwing it up to not expose a potential problem is asking for something to go wrong. Humans can/will act STUPIDLY. For proof, go to www.darwinawards.com
.So even if you think it will never be encountered Jamie, rest asured that some moron will do it. And even though it would be nice to see that person's genes removed form the gene pool, it is still not a good thought that someone could possibly die or be severely injured from a small software bug.
If it is real, it should be fixed. There should be no arguing that point.
Brules
I never argued that it shouldnt be fixed it it does exist. We still don't know if it does exist in an actual dive scenario, only that a few people have seen it in SIM mode, which according to my above post, doesn't necessarily mean squat. If someone gets toxed in SIM mode, who cares as long as the actual dive calc is correct, which hasn't been determined for sure.
Relying on a dive computer unconditionally is asking for a problem... A person is 100% responsible for planning the dive and not screwing up... why? Because its their life and possibly the lives of their dive buddies. We have the ability to recognize our mistakes guys, the dive computers don't. If you truly plan a dive, there are not mistakes. If mistakes are made, you only half planned it. Planning and checking the plan go hand in hand. Plan the Dive, Dive the Plan. Do not rely on your dive computer to think for you. Where the heck is UP?
And since after reading that last paragraph, you guys will say once again that I'm saying that suunto shouldn't fix the problem... yes, suunto should fix any problem that doesn indeed exist in their dive computers. Suunto quickly corrected the problem with the Vytec, why would they have avoid it with the Vyper? The Vytec is more expensive and they replaced them, why not replace the Vyper if its has a problem?
I look forward to hearing Suunto's response and offer this challenge. Those of you that say its such a conservative dive, strap on 32% and do the dive and then start collecting some hard evidence. Print the graphs and it will show the OLF. If you won't do the dive, why not? Because its too risky?
Relying on a dive computer unconditionally is asking for a problem... A person is 100% responsible for planning the dive and not screwing up... why? Because its their life and possibly the lives of their dive buddies. We have the ability to recognize our mistakes guys, the dive computers don't. If you truly plan a dive, there are not mistakes. If mistakes are made, you only half planned it. Planning and checking the plan go hand in hand. Plan the Dive, Dive the Plan. Do not rely on your dive computer to think for you. Where the heck is UP?
And since after reading that last paragraph, you guys will say once again that I'm saying that suunto shouldn't fix the problem... yes, suunto should fix any problem that doesn indeed exist in their dive computers. Suunto quickly corrected the problem with the Vytec, why would they have avoid it with the Vyper? The Vytec is more expensive and they replaced them, why not replace the Vyper if its has a problem?
I look forward to hearing Suunto's response and offer this challenge. Those of you that say its such a conservative dive, strap on 32% and do the dive and then start collecting some hard evidence. Print the graphs and it will show the OLF. If you won't do the dive, why not? Because its too risky?
Aquamaniac
Contributor
- Messages
- 1,257
- Reaction score
- 5
This one is wide open Ron............
Similar threads
