M-DCT Lag - What's the reason?

[footie]

So it appears we have someone on version 33.0 (new software fix) and still the lag is present in all it's glory.

At what point will the optimists (swamp included) conceed that I was indeed correct and it's basic design has the flaw and no amount of upgrades will cure or eliminate it completely.

[Radiation Joe]

Originally Posted by footie

So it appears we have someone on version 33.0 (new software fix) and still the lag is present in all it's glory.

At what point will the optimists (swamp included) conceed that I was indeed correct and it's basic design has the flaw and no amount of upgrades will cure or eliminate it completely.

I've always been afraid of that, but am holding out hope that it can be fixed.

[gr8000]

Originally Posted by footie

So it appears we have someone on version 33.0 (new software fix) and still the lag is present in all it's glory.

At what point will the optimists (swamp included) conceed that I was indeed correct and it's basic design has the flaw and no amount of upgrades will cure or eliminate it completely.

I don't believe this for 3 main reasons:
1. It was not there before (at least not to the extent it is now)
2. This issue does not appear in downshifting to 1st gear (which is a much more "difficult" gear to handle).
3. Common sence. I am not a DCT specialist knowing how the DCT was designed (far from that) but knowing the very basics, in my logic I cannot convince myself why the lag cannot be fixed.

Anyway, let's see if and when the lag will be fixed... I will chase BMW with another follow up letter to see what the official position is.

[mkoesel]

Originally Posted by footie

At what point will the optimists (swamp included) conceed that I was indeed correct and it's basic design has the flaw and no amount of upgrades will cure or eliminate it completely.

Footie, you of all people should remember that this lag was made much more severe by the last update. It is clearly not a hardware flaw. Perhaps there will always be some lag there, but clearly software can control the magnitude of it.

Also, we have no proof yet that PROGMAN 33.0 even included an official fix for the lag. I realize that nitramsen has an "in" of some sort, but I'd like verification. Plus he mentioned something about a "custom version". If this update was indeed meant for the masses and meant to correct something of significance, then we should get much more information (though nothing of substance from official sources) as the update is propigates and more is discovered about it. Personally I am skeptical any of that will happen, because.

[footie]

Originally Posted by mkoesel

Footie, you of all people should remember that this lag was made much more severe by the last update. It is clearly not a hardware flaw. Perhaps there will always be some lag there, but clearly software can control the magnitude of it.

I am basically saying the same thing, that is that the lag maybe be controlled by software fixes but never to the point where it's completely gone.

Now in my opinion that would mean that it DOES have an underlining flaw in it's design for this to not have been fixed, whether the flaw is lag or something else but with each cure there's something else that seems to get worse.

Originally Posted by mkoesel

Also, we have no proof yet that PROGMAN 33.0 even included an official fix for the lag. I realize that nitramsen has an "in" of some sort, but I'd like verification. Plus he mentioned something about a "custom version". If this update was indeed meant for the masses and meant to correct something of significance, then we should get much more information (though nothing of substance from official sources) as the update is propigates and more is discovered about it. Personally I am skeptical any of that will happen, because.

Again that is why I said the words 'it appears' but until we know for sure we are guessing. But I am very unhappy with BMW that after almost one year of production and countless more of development that we are having issues with this gearbox.

I have driven a few DCT gearboxes now and the only one with these issues is the M-DCT, in fact compared to Porsche's PDK the M-DCT isn't half the polished package which to me is proof positive that BMW released their version far to soon and with not enough development work.

Here's a question for you. Does the Ferrari DCT suffer the same issues?

[mkoesel]

Originally Posted by footie

I am basically saying the same thing, that is that the lag maybe be controlled by software fixes but never to the point where it's completely gone.

Ok, but arguing that point in the context of this new update does not make much sense. We already know that the new update does not make the lag better than the original software so the two things (the lag and this update) have no relevance to each other whatsoever.

Now in my opinion that would mean that it DOES have an underlining flaw in it's design for this to not have been fixed, whether the flaw is lag or something else but with each cure there's something else that seems to get worse.

If the lag can be made short enough to where it does not hinder driving, then it can no longer be considered a flaw, whether it be a hardware or software one. I believe this is possible. But no evidence exists to prove nor disprove that yet. Nothing has really changed, and all we are doing is adding to the volumes of speculation that already exist on the topic.

Again that is why I said the words 'it appears' but until we know for sure we are guessing.

You did say that, but you also issued a query to the "believers" as to when we might concede the flaw for good. I don't think there is anything new that will sway anyone (from either side).

Here's a question for you. Does the Ferrari DCT suffer the same issues?

No idea, never driven any Ferrari.

No one is arguing that other DCT products are able to operate without the low speed lag. No one is arguing whether the BMW software has bugs. So citing comparable examples, again, isn't adding anything new right now. We've discussed all of this before, and we agree that the lag is an issue (a serious one) that other comparable transmissions do not exhibit. I just don't personally think it is unsolveable via software at this point. And frankly if it is, then BMW is in for a long road because this will eventually get the attention of the NHTSA (it could be years, who knows) and they will eventually force BMW and their supplier to fix all the cars (at least in the US) even if it means redeveloping the entire transmission from scratch. Unfortunately, it may take death or serious injury for that to happen.

[swamp2]

Originally Posted by footie

I am basically saying the same thing, that is that the lag maybe be controlled by software fixes but never to the point where it's completely gone.

Hmmm, it doesn't seem that you, mkoesel and I are saying the same thing at all. You are purporting there are essentially unfixable design flaws causing the lag. Most others, based primarily on the behavior of the tranny with the first software revision from the factory, claim it is primarily a software only issue that was altered significantly and for the worse with the first software revision. It seems painfully obvious that it can be fixed. However, there is the remote possibility that this lag is intricately linked to the previous delay between paddle actuation and the beginning of clutching. Others have suggested this but I find the two only very, very loosely related.

[footie]

I am basically saying that the lag may be the resulting best solution of curing the other issues that where present in the transmission. Maybe I am the sole person here that finds it amazing that M-DCT was launched with so many problems and is still having problems after (if memory serves me) at least two upgrades.

There is many things that BMW have got right with the transmission, D4 is one setting that is very well judged for spirited driving while still proving useful around town, S1-2 is also a great balance between speed of shift and smoothness and finally D1 is very good in snow. Another great thing is the gear stick and how it's used, other should follow their example in my opinion. But for a car that is so well developed and judged else where it highlights the gearbox's failings all the more.

[mkoesel]

Originally Posted by footie

I am basically saying that the lag may be the resulting best solution of curing the other issues that where present in the transmission.

That possibility remains footie but this latest (and frankly, still mysterious) update does nothing to convince me.

What would hold some weight to sway my opinion is if they released an update that very clearly addresses the low speed lag (fixes it or makes it much better), but then recreates old bugs fixed in previous releases. I would then begin to wonder if the transmission just cannot be "balanced" for lack of better word to where all systems operate harmoniously together and "all holes are plugged" (for lack of better idiom).

But that hasn't happened yet. So again, really there is nothing new to glean from this latest update.

Maybe I am the sole person here that finds it amazing that M-DCT was launched with so many problems and is still having problems after (if memory serves me) at least two upgrades.

I am not so surprised at the bugs. In fact, there was discussion of software updates being a strong liklihood even before the transmission debuted. All things considered, the transmission does its primary job (to change gear ratios really freakin fast) very well. You echoed this sentiment in your post too, so on that much we agree. Clearly BMW spent a lot of time on that aspect, as they should. The rest is minor, at least in principle. Was the transmission on an accelerated development timeline? Maybe - there is some (circumstantial) evidence to suggest it. What is unfortunate is how BMW is handling the lag situation. Sure there are probably scads of lawyers involved whose job it is to protect the company, but I still think the PR could be much, much better. If someone dies, man, I would not want to be one of the guys who put in all the hard work on this transmission. It would really give the whole thing a serious black eye, and that'd be a shame because it is mostly (99%) a really great product.

[footie]

This might be worth checking but you now get the DCT in the 335, admittedly with a different software with less setting, is it known whether the lag is present on it also.

[southlight]

Originally Posted by footie

This might be worth checking but you now get the DCT in the 335, admittedly with a different software with less setting, is it known whether the lag is present on it also.

It's not even sure that this version of DCT is being supplied by the same manufacturer.


Best regards, south

[footie]

Originally Posted by southlight

It's not even sure that this version of DCT is being supplied by the same manufacturer.


Best regards, south

If that is true then sure that strengthens my case that the M-DCT is flawed, why else would they use another gearbox supplier when a perfectly good one has already been developed, the only requirement would be scaling down the options on the software.

[valeram]

Originally Posted by footie

If that is true then sure that strengthens my case that the M-DCT is flawed, why else would they use another gearbox supplier when a perfectly good one has already been developed, the only requirement would be scaling down the options on the software.

Everything that was said was just a hunch. What we know for sure that the normal 3 series DCT will have less programs compared to the M-DCT (based from BMW announcement).

[footie]

Originally Posted by valeram

Everything that was said was just a hunch. What we know for sure that the normal 3 series DCT will have less programs compared to the M-DCT (based from BMW announcement).

The 335 DCT is already out in Germany and the likes, so surely someone has driven one and can tell us whether the lag is present there too or not.

P.S.
South is usually right on these things, so if he reckons it's different then chances are it is.

[southlight]

Originally Posted by footie

The 335 DCT is already out in Germany and the likes, so surely someone has driven one and can tell us whether the lag is present there too or not.

P.S.
South is usually right on these things, so if he reckons it's different then chances are it is.

Seems like I was wrong. Realoem lists the same transmission (GS7D36SG) for both cars, but shows different part numbers. So the odds are high that you're right and only the software (control unit) is different. This comes surprising as I thought an M transmission is more sophisticated, thus more expensive than the AG counterpart.


Best regards, south

[mkoesel]

Originally Posted by footie

Why else would they use another gearbox supplier when a perfectly good one has already been developed, the only requirement would be scaling down the options on the software.

- Cost
- Complexity
- Availability

And probably a half dozen other possibilities I didn't think of. Footie, your witch hunt against the M-DCT transmission is causing you all sorts of irrational conclusions. The 335i DCT does not use the same paddle arrangement as the M3 M-DCT either. So then, we must conclude the M3 paddles are flawed also, yes? I mean, why else would they use a different setup on the 335i?

[mkoesel]

Originally Posted by southlight

This comes surprising as I thought an M transmission is more sophisticated, thus more expensive than the AG counterpart.

South, seems to me then, it is very similar to the manual transmissions in these two cars. Same part, although the M gets an additional cooling unit as I recall it.

[footie]

Originally Posted by southlight

Seems like I was wrong. Realoem lists the same transmission (GS7D36SG) for both cars, but shows different part numbers. So the odds are high that you're right and only the software (control unit) is different. This comes surprising as I thought an M transmission is more sophisticated, thus more expensive than the AG counterpart.


Best regards, south

South it's no surprise, VAG has been doing the same thing for years. An example with Cruise Control in the Golf and A3, in the Mk1 A3 this extra was near double the identical unit in the Golf and only recently has the prices come into line.

Getting back to the gearbox, if it does prove to be lag free then it may mean the complexity of the software is the M-DCT's downfall and I was wrong to consider that the gearbox design was flawed.

Though this is yet to be proven either way.

[footie]

Originally Posted by mkoesel

- Cost
- Complexity
- Availability

And probably a half dozen other possibilities I didn't think of. Footie, your witch hunt against the M-DCT transmission is causing you all sorts of irrational conclusions. The 335i DCT does not use the same paddle arrangement as the M3 M-DCT either. So then, we must conclude the M3 paddles are flawed also, yes? I mean, why else would they use a different setup on the 335i?

It's not a witch hunt, I am a little upset as many are that after so long the gearbox is still causing problems. Given BMW's M-division's engineering credentials I expected more, and given that it's the dearest of all the DCTs from rival competitors I think it should have be properly developed.

As for the paddle differences, I don't understand why BMW don't offer the M3 style paddles as the other ones are no better than the shit Porsche put with their PDK.

Who here would have been happier with only 3 manual and 2 auto modes if it worked 100%, I know I would if that happened to be the cure. Complexity leads to problems no matter what it is.

[mkoesel]

Originally Posted by footie

It's not a witch hunt, I am a little upset as many are that after so long the gearbox is still causing problems.

Right, and my point is that your emotions are causing your to make some nonsensical claims. I'm not trying to be a jerk here, but honestly, you may wish to just clear your head before leveling any more assertions about the transmission. I know you are capable of rational thought (your contributions here are traditionally high quality and welcome) but in this instance I believe you are letting your frustrations get the best of you.

As for the paddle differences, I don't understand why BMW don't offer the M3 style paddles as the other ones are no better than the shit Porsche put with their PDK.

It is likely one of or a combination of all those I list above.

Who here would have been happier with only 3 manual and 2 auto modes if it worked 100%, I know I would if that happened to be the cure. Complexity leads to problems no matter what it is.

I would be happy as long as the five modes were equivalent to S1, S5, S6, D1 and D5. I do agree that the extra modes, for me at least, are pretty much superfluous and I don't need them.

However, again here, even if we find that the 335i implementation is lag free, it still can not be concluded 100% that the additional modes in the M3 are the cause for the M3 issues. The M3 has a completely different drivetrain including the engine and differential, not to mention all the electronics that control it. The complex interaction between the transmission and these things means that the two will likely be programmed very differently regardless of how many Drivelogic programs are involved.

[swamp2]

Originally Posted by southlight

Seems like I was wrong. Realoem lists the same transmission (GS7D36SG) for both cars, but shows different part numbers. So the odds are high that you're right and only the software (control unit) is different. This comes surprising as I thought an M transmission is more sophisticated, thus more expensive than the AG counterpart.

Makes perfect sense to me. I always make arguments here bases on the economy of scale. Here the cars have almost identical torque output AND have the same basic chassis. Using the same gearbox with different software makes perfect sense. I alluded to this happening a long time ago - DCT customizations solely via software. Like footie I would love to know of the 335 with DCT has this particular lag. If not it would put it to rest quite well that the box is not flawed - just the software.

South - can you ask around on some forums in Europe about this? An extra big thanks in advance if you can (especially given your club membership ).

[footie]

Originally Posted by mkoesel

I would be happy as long as the five modes were equivalent to S1, S5, S6, D1 and D5. I do agree that the extra modes, for me at least, are pretty much superfluous and I don't need them.

However, again here, even if we find that the 335i implementation is lag free, it still can not be concluded 100% that the additional modes in the M3 are the cause for the M3 issues. The M3 has a completely different drivetrain including the engine and differential, not to mention all the electronics that control it. The complex interaction between the transmission and these things means that the two will likely be programmed very differently regardless of how many Drivelogic programs are involved.

The difference in engine and drivetrain should have nothing to do with the lag experienced, VAG have used their DSG on numerous engines ranging from diesels right up to the TT-S and yet all of them don't suffer or behave any different.

My argument is that IF (big if) but if the DCT in the 335 is indeed lag free then the software setup in the M3 is seriously flawed for one reason or another. But none of this matters, the problem I have is the supposed years of development done by BMW plus the extra year of production and still we have these issues with the gearbox.

I personally expected more from M-Division and to be honest with you, I thought you would have felt the same way. Yeah it's frustrating but it can be a serious problem, especially as we in Europe have lots of roundabouts (rotors) where you merge with moving traffic on a regular basis.