If it is consistently 1000rpm rather than a % could it be that Ducati have done a deal with Dorna to manipulate the on board so as not to leak their technological advantage and displacement decisions? I think it would be perfectly reasonable for Ducati to request such a thing.
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
Could be. Something like that is certainly possible.
Or perhaps the bikes only broadcast RPM as a % of peak, not an absolute value? This sounds odd, but from a telemetry POV it might make some sense(?) If the Dorna computer that generates the on-screen tach has the wrong scale factor (someone mis-configured something...) , it will obviously produce garbage values.
Keep digging!
You make a good point about the integrity of the video b/c youtube isn't exactly refereed data. However, the video is a compilation of various onboard shots. The 4 samples were taken from two different onboard videos (combined into a single youtube compilation). To check the integrity of the entire video, I spot checked a few locations
Sampled Yamaha @ 1:48.3 (size 8192), observed 151hz (18,120rpm) compared to 18,246rpm onboard
Sampled Honda @ 2:36.5 (size 8192), observed 158hz (18,960rpm) compared to 18,869rpm onboard
Sampled Honda @ 5:28.5 (size 4096), observed 156hz (18,720rpm) compared to 18,724rpm onboard
The video is at least consistent. Dorna's rev counter could be off b/c the Ducati onboard footage has lower rev ceilings than the other 800s by about 1000rpm. Either way this is a weird situation. Ducati said they were going to build a bike around 930cc, but they claim to have built the GP12 at 1000cc. Dorna might not be able to correctly display Ducati's actual rev ceiling.
I don't want to throw your baby out, but we are sort of in no man's land right now. If Krop publishes, I think it should be "Is the Ducati a 930cc??Maybe??Perhaps??"
If I can find more Ducati onboard footage, maybe from a different year, we might have a better idea. Any MotoGP.com members who can lend a hand?
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
Maybe I've missed something.
As I understand, we have one frankenvideo from 2010/2011? in which the Duc's RPM, as determined by manual analysis or spectrum peaks (which match) doesn't agree with the on-screen Tach. That's certainly an anomaly in need of an answer, but I don't see it invalidating our analysis of the 2012 Duc, or the techniques as a whole.
FWIW, I've just looked again at several points on the 2012 Rossi Onboard and am (still) seeing 17,5 and 17,6 at multiple shift points. This is ~1300 RPM faster than the highest Yam or Honda revs yet recorded.
And, as you point out, the numbers align perfectly. 16350 / 17600 = .929 = 930cc
(16350 is Yam record to date - Spies Qual)
I still say something is up. Even with room for bad video, etc., there is cause for suspicion. (Don't forget the long-standing rumors of 9xx.) If Krop is unconvinced, he can still go
and ask Prezi, "The GP12 revs 1300 higher than any of your competitors. How can it do that?
"
As I understand, we have one frankenvideo from 2010/2011? in which the Duc's RPM, as determined by manual analysis or spectrum peaks (which match) doesn't agree with the on-screen Tach. That's certainly an anomaly in need of an answer, but I don't see it invalidating our analysis of the 2012 Duc, or the techniques as a whole.
FWIW, I've just looked again at several points on the 2012 Rossi Onboard and am (still) seeing 17,5 and 17,6 at multiple shift points. This is ~1300 RPM faster than the highest Yam or Honda revs yet recorded.
And, as you point out, the numbers align perfectly. 16350 / 17600 = .929 = 930cc
(16350 is Yam record to date - Spies Qual)
I still say something is up. Even with room for bad video, etc., there is cause for suspicion. (Don't forget the long-standing rumors of 9xx.) If Krop is unconvinced, he can still go
The reality is that we have no control data, and by discarding conflicting data points, we are implying that Dorna cannot install rev counters on big bang Ducati GP bikes. We've only amassed a pile of circumstantial evidence, and a basic (inherently sound theory, imo) about counting amplitude waves over a period of time. It's not a bad situation, but it doesn't overcome an anomaly like official race footage where the numbers don't match up.
Anyway, I found some onboard footage from Hayden's bike. 2011 Brno Qualifying session with 23:30 left on the official timer (so 37:30 into the session).
At 23:39 during 4th gear acceleration, peak revs observed 17,609rpm. The spectrum analysis is a mess, but it shows a small peak at 148hz (17,760rpm). Same run in 5th gear, peak revs observed 17,828. Spectrum analysis, another mess, did show a peak at 149hz (17,880rpm). Same run in 6th gear, peak revs observed 17,442rpm. Spectrum analysis says 144hz (17,280rpm).
I think we can safely say that the spectrum analyzer can replicate results, and that Dorna may have had issues with the onboard Ducati equipment in 2010 (either that or the Ducati rips were bad). I'm going to keep searching for footage b/c the commentators were yucking it up during Hayden's onboard.
Anyway, I found some onboard footage from Hayden's bike. 2011 Brno Qualifying session with 23:30 left on the official timer (so 37:30 into the session).
At 23:39 during 4th gear acceleration, peak revs observed 17,609rpm. The spectrum analysis is a mess, but it shows a small peak at 148hz (17,760rpm). Same run in 5th gear, peak revs observed 17,828. Spectrum analysis, another mess, did show a peak at 149hz (17,880rpm). Same run in 6th gear, peak revs observed 17,442rpm. Spectrum analysis says 144hz (17,280rpm).
I think we can safely say that the spectrum analyzer can replicate results, and that Dorna may have had issues with the onboard Ducati equipment in 2010 (either that or the Ducati rips were bad). I'm going to keep searching for footage b/c the commentators were yucking it up during Hayden's onboard.
Joined Jun 2007
10K Posts | 1K+
Norah Head
Look at the Duc waveform. The peak-to-peak indicates 17k
Let's not throw the baby in the dumpster just yet!
I can't think of a good reason the revs wound be a constant 1000PRM too high. If there was a problem, you'd expect a proportional scale factor issue, not a static offset. Curious.
I looked at the last sample, at 8.29.6 and got this:
13485:8296Duc.jpg]
Manual analysis backs up your frequency sniffer - the spreadsheet says 18,866 RPM.
Either the video has been sped up, or the Dorna RPM-O-Meter is wacky. I don't yet see a proven problem with either the manual analysis or spectrum-peak-HZ methods.
Run the Audacity sniffer on the attached WAV file, and please manually look at the peak-peak time interval.
always check with a manual analysis. You can check that it is working via the spectrum plot by cutting a fairly constant section, that is easy to manually decipher then check that against the frequency analyzer.
To be honest I see the manual perusal and calcs as the more correct method, The spectrum analyzer I would use to merely guide yourself to what you are possibly expecting.
Joined Jun 2007
10K Posts | 1K+
Norah Head
Maybe I've missed something.
As I understand, we have one frankenvideo from 2010/2011? in which the Duc's RPM, as determined by manual analysis or spectrum peaks (which match) doesn't agree with the on-screen Tach. That's certainly an anomaly in need of an answer, but I don't see it invalidating our analysis of the 2012 Duc, or the techniques as a whole.
FWIW, I've just looked again at several points on the 2012 Rossi Onboard and am (still) seeing 17,5 and 17,6 at multiple shift points. This is ~1300 RPM faster than the highest Yam or Honda revs yet recorded.
And, as you point out, the numbers align perfectly. 16350 / 17600 = .929 = 930cc
(16350 is Yam record to date - Spies Qual)
I still say something is up. Even with room for bad video, etc., there is cause for suspicion. (Don't forget the long-standing rumors of 9xx.) If Krop is unconvinced, he can still goand ask Prezi, "The GP12 revs 1300 higher than any of your competitors. How can it do that?"
WHoa ..... those onscreen tachs are likely pretty out! they are a minimal glimpse ( I assume they have a minimal window of data to use hence are of very low resolution with a wandering latency ). Could be they are quite out with the section of the sound curve they are representing by being displayed on screen.
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
Indeed.
Particularly if there are several adjacent frequency peaks, or if the sound is low quality, or if a particular engine is 'new' to you.
You can't consider all Dorna-generated sound files, ripped from youtube, as gospel truth, and then discard onscreen graphics when convenient. There is no science behind discarding outlying points without a compelling reason. We are workaday amateurs, armed with freeware, using the same metering techniques for amplitude wave measurement that we learned in high school applied physics. The GPC is a collection of competent executives regarding technical matters, like weights, measurements, and mechanical engineering (including entertainment engineering). If we are going to assume incompetence by one party without any evidence, we should probably start with us.
When I rip the GPC, I generally claim that they have no leadership intangibles, no vision, a poor understanding of the economics of government, and low intuitive capabilities. They prove it everyday by leading the sport down the road to economic contraction, and by grafting technical paradigms and economic paradigms, from other racing series, onto MotoGP without examining the unintended consequences. You claim that Dorna can produce distortion-free onboard audio, but they can't create a rev-counter (probably with the help of the MSMA). You'll need more than assumptions.
Geo's work is good, and it is probably correct, but their are some strange outlying points, sourced via the same method we used to source the Ducati onboard footage from Qatar. If we have a control, we don't have to worry about outlying points. Right now we are merely backing into rpm with hertz, and backing into hertz with a time tested, though not always precise, method of counting amplitude wave periods. We assume that the people who provided the data are incompetent when convenient, but also competent when convenient. We should probably find more Ducati onboard footage.
Joined Oct 2006
25K Posts | 4K+
Your Mom's House
Lex, i think the 'sane' people following thread were/are aware and should be under no illusions, but thanks for spelling it out. Obviously primary source will be best, but that doesnt/shouldnt stop us from exploring and hypothesizing using clever means and the rules of physics (which are constant at any level).
I think Geo's work will prove correct, but I'm still concerned by the anomaly in the onboard videos I tested, which worked flawlessly for the Honda and Yamaha, but not for the Ducati. A secondary test on Ducati onboard footage from 2011 was relatively successful.
Establishing control data won't change whether Geo is correct, but it does seem that if we start playing scientist, we are obliged to follow through on some sort of scientific method. I will see if I can find more Ducati onboard sometime later today. If Krop posts Geo's findings, it should probably be sooner rather than later, in case Dorna show up with a rev counter at Jerez, and Geo's discovery becomes common knowledge.
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
Dude, you really do see everything in stark black-and-white, don't you.
There's nothing inherent in the Ducati waveform that makes analysis harder or less accurate. There is a very distinct, repeating pattern that is actually much 'cleaner' than many of the other bikes. I think the Yamaha is the messiest, with few clear spikes and tons of harmonics bouncing about.
At the same time, you do have a point. To date, we have only two 2012 Duc samples.
1 The 2 minute MotoGP on-board that indicates 17,5+
2 The onboard race video, where I saw 16,4 and 16,8 from Rossi's bike, taken from the only two decent halfway decent Duc onboard samples available.
More onboard Duc is certainly a priority, but for now I see no reason to toss the 17,5xx data "without a compelling reason."
I might add that 16.8 from a potentially cruising Rossi is still 500 RPM higher than anything else on that video.
Lex, sincere thanks for the help and contrarian view. (We don't want to become an army of blind Boppers, do we?
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
I have a sneaky suspicion that an on-screen Duc Tach is not going to happen until the 'new' engine arrives. Conspiracy abounds!
So far, the only 2012 tach I've seen was on the Honda. OTOH, I've not been looking for one on either Japanese bike, so might have missed a Yam appearance.
It doesn't bother you that we can reproduce Dorna's results for the Honda and the Yamaha within 1%, but not for the Ducati (not conclusively anyway)?
It's not necessarily a deal breaker b/c wave analysis is quite solid, but it couldn't hurt to double check with more onboard footage from 2011.
Joined Jun 2007
10K Posts | 1K+
Norah Head
I hope you are taking the micky??
The tacho provided on screen from Dorna could be equated to Geo giving us a result for his calculations only ....... no reference to data or method shown.
Surely you are not suggesting what is being deduced from the wav files is Gospel?? I'm pretty sure most of us know that it is a "possible" ...... nothing more. Though given that these software devices are pretty accurate, repeatable and verifiable, I would suggest they aren't that far off.
The "TV tacho" is possibly a guy sitting there with a slot car throttle listening to the sound of the bike and depressing the trigger in conjunction with an increase in revs he hears ........... who knows
EDIT:nothing new...
started replying from the beginning of the thread, saw it was all covered
as you were...
as you were...
Joined May 2008
3K Posts | 79+
Southern Aridzona - Hell's doorstep.
.
JEREZ
Youtube MotoGP just posted 3 new onboard vids from Jerez.
It looks like one of the practice sessions, with a rather damp track.
Audio level is low, and some of the samples are a big dodgy. The Duc is actually the easiest to hand analyze; it has a nice, easy to spot, repeating pattern. The other two are more chaotic and I occasionally found myself leaning on the spectrum sniffer to determine likely wave pattern periods. Once identified, these weaker patterns are, IMO, good enough to verify the spectrum values. Still, there is a bit of circular logic at work with the Yam and Honda numbers.
Duc RPM: 16,680 16,560 16,540 16,800
http://www.youtube.com/watch?v=6KG4oyTXeLw
Yam RPM: 13,800 14,100 14,600 15,500 14,200 15,600
http://www.youtube.com/watch?v=Gg9NN89BJdc
Honda RPM: 16,300 15,700 15,900 16,100
http://www.youtube.com/watch?v=r3GzCEtUN-Q
JEREZ
Youtube MotoGP just posted 3 new onboard vids from Jerez.
It looks like one of the practice sessions, with a rather damp track.
Audio level is low, and some of the samples are a big dodgy. The Duc is actually the easiest to hand analyze; it has a nice, easy to spot, repeating pattern. The other two are more chaotic and I occasionally found myself leaning on the spectrum sniffer to determine likely wave pattern periods. Once identified, these weaker patterns are, IMO, good enough to verify the spectrum values. Still, there is a bit of circular logic at work with the Yam and Honda numbers.
Duc RPM: 16,680 16,560 16,540 16,800
http://www.youtube.com/watch?v=6KG4oyTXeLw
Yam RPM: 13,800 14,100 14,600 15,500 14,200 15,600
http://www.youtube.com/watch?v=Gg9NN89BJdc
Honda RPM: 16,300 15,700 15,900 16,100
http://www.youtube.com/watch?v=r3GzCEtUN-Q
Joined Jan 2011
3K Posts | 152+
The North!
OMG! You guys screw with my head! I have no idea what your on about or how your getting your readings/outcomes but I find it so interesting reading what your findings are. Even if I don't understand them!
WTF???
WTF???
