Login required to started new threads

Login required to post replies

Prev Next
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [trail] [ In reply to ]
Quote | Reply
Here's another wrinkle: I just looked back at a "yaw-weighted average drag" of Mavic's new "Cosmic Ultimate" wheels (which are pretty sweet btw) and noticed something unusual about the yaw weighting. Take a look:



See it? I'm not talking about the unusually good Aeolus 3 performance. Take a good long stare. Is it me or is Mavic placing an unusual emphasis between 10 and 20 degrees of yaw (and not for the sake of winning their own tunnel shootout it would seem)? Doing rough approximations, it does appear that the final value they give for each wheel weights primarily measurements inside of 10 degrees of yaw but... what's with all of those data points out past 10 degrees?

I'd also like to point out that, despite some tongue-in-cheek marketing, Mavic has been pretty tenacious in sticking with NACA derived shapes.
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [GreenPlease] [ In reply to ]
Quote | Reply
When close to the stall angle, the data tend to become more variable, as very small differences in yaw angle result in much larger changes in drag. It therefore makes sense that they have collected more data between 10 and 20 deg - in fact, I did exactly the same thing when testing the flat-faced MRC brake in my wind tunnel, collecting data at 12.5, as well as 10 and 15, deg of yaw to try to better understand the source of asymmetry.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [kriss] [ In reply to ]
Quote | Reply
They come with black prince pads, and all the items to go tubeless. They are a great value. I never even think about the braking. They just work. I have a set on my trek now. The tubeless part is a PIA. Ran 24 Bontrager tires and it took 2 of us to get on. The rear seated great and has not leaked. The front looses a little in a week.

http://www.TriScottsdale.org
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [GreenPlease] [ In reply to ]
Quote | Reply
GreenPlease wrote:
<snip> post #1 </snip>


Great topic - and I'm glad to see folks challenging the status quo when it comes to how best evaluate wheels in a tunnel. The vertical and horizontal gust simulations (louvers) that are alluded to by the folks doing the testing is really interesting! I'd like to hear more about this. 👍

Side note: glad to see that post #2 in this thread related to tires! Warms my heart! :-) I can still remember the shock on the Rolf wheels aerodynamic consultant in the 90's when I put skinny tires on wheels that were being tested in the tunnel at A&M and minds were seemingly blown! haha.

Now, I just wish post #3 would have something to do with "watts to spin"! haha. ;-)

Unsteady aerodynamics are very interesting to me => here are my thoughts on the topic from 12+ years ago:

http://biketechreview.com/...adence-and-cda#11354


biketechreview wrote:

IMHO (and that goes for this entire post), bicycles experience quasi-static flow conditions for the majority of their use. This is to say that the speed of the maneuvers (rocking/steering/leaning/pedalling etc.) occur too slowly for the flow to be considered truly dynamic in nature. The flowfield on the body has reached a quasi steady state before the orientation of the body can change yet again - a particle in the flow will pass over the entire length of the body before the body can change its orientation significantly.

If one wants math, the quantities of interest are t_hat (non dimensional maneuver time) and t_star (t_star = L/V -> the time needed for the flow to pass over the body). In order for the flow to be considered dynamic, and thus yield wind tunnel testing in a static environment invalid, t_hat generally must be less than ~3 (if my memory serves me right).

t_hat = t/t_star=(t*V)/L

where t is maneuver time, V is flow speed, and L is the length scale (chord length of a fork blade, rim depth, torso length, leg diameter etc.).

For bike _parts_, I simply don't think that maneuvers happen fast enough (nor are the length scales long enough) to make the quasi-static assumption break down - it is also assumed that beta angle sweeps in the tunnel are done for all cases one is trying to model. For the rider and her position, it becomes a possibility that the flow is on the verge of becoming fully dynamic (even then, I doubt it - though I don't know for sure), and only than at the lower speeds and only for the torso length scale & not leg diameters.

you'll note the caveat in the quote above: " it is also assumed that beta angle sweeps in the tunnel are done for all cases one is trying to model"...i think there is a dearth of information/tunnel data regarding a wheel that is simultaneously experiencing sideslip and "roll" => maybe this protocol can provide insight into this?

anyway, good stuff - I really appreciate that a group is trying to move the needle on the topic!

=================
Kraig Willett
http://www.biketechreview.com - check out our reduced report pricing
=================
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [GreenPlease] [ In reply to ]
Quote | Reply
GreenPlease wrote:
That said, rims like the Enve 7.8 and 808 are so different in shape but so similar in performance it leads me to believe there’s more than one viable approach.

i recall a comment from Paul Lew about the original Reynolds Aero line a few years ago basically saying (as best i recall and i'm sure i'm mixing up the terminology but hopefully i can convey the point) that the more V shaped rims are low drag whereas the more U shaped are high lift (negative drag sail effect) - producing opposite forces along different vectors such that the resolved vector sum may amount to much the same system performance. of course the different shapes have pros and cons in terms of rim structure and handling which is mentioned in the Princeton thread

i found this - note the quote i was thinking of but along the same lines
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [pk1] [ In reply to ]
Quote | Reply
The best way to think of this is that for a given depth of wheel, they are all going to have really similar surface areas... so the amount of total force that has to be resolved is going to be somewhat similar. This is how the whole Firecrest thing came about... once we had deep wheels that could output drag numbers nearly identical to a disc, we believed that there would be little more speed to be found, but that we should focus on handling as the real problem with deep wheels was that in the conditions where they were most advantageous to run (in the wind) people didn't want to ride them.. So all of the surface of the wheel has some forces on it, the question is how do you resolve and distribute them to minimize drag, side force and steering torque. Ultimately we learned that it becomes a bit of a shell game where every possible solution has some tradeoff and once the system is highly optimized the tradeoffs generally become nearly 1:1 where 10 grams of drag reduction just becomes 10 grams of side force, etc..

I don't have access to all my old Zipp stuff, but here's a link to a presentation Matt Godo gave in 2011 using our 2009-2010 benchmarking study for Firecrest.. there are other presentations that show not only steering torque (he called it turning moment) but wattage to spin as well as the transient effects of turbulent and vortex shedding. Also shown here but only minimally are the visualization tools we developed to understand where on the wheels the forces were acting which allowed the calculations of steering torques and centers of pressure (Cp) which we determined to be the biggest factor in handling.

Once you eliminate the low hanging fruit of big pressure drags related to separations of flow, the only way left to sufficiently lower the drag further is to utilize aerodynamic forces have to be turned into a combination of side lift and forward lift.. The goal with Firecrest was to balance the front and rear half of the wheel so that the center of pressure remained in front of the steering axis at all angles with little fore-aft movement. You see on the graph that a front disc wheel and our own 1080 have a rearward center of pressure..and while this 'rudder effect' seems initially like a self stabilizing feature, it isn't. This is due to the way a bicycle initiates a turn.. for a bicycle to turn left, the contact patch has to move to the right, so the wheel initially has to turn right which then makes the bike lean left... this is why you fall over when you get pinned up against a curb or in a rut, you can't lean away from the curb because the wheel needs to go in the curb direction to move the contact patch to allow you to lean the other way.

Similarly, we found with the wheels..if a wheel turns into a gust wind (sounds like a good idea) the contact patch moves toward the wind, but the rider lean then goes the other way which pushes the bike/rider system along the direction of the wind and the only way to correct it is a pretty significant steer into the lean direction to move the contact patch to the other side... with Cp in front of the steering axis, a gust of wind initiates a lean/turn back into the wind.. it's much more stable.

So the problem turns out that some wheel designs, particularly ones that have sharper inner diameters is that the center of pressure will cross the steering axis at some angle..as you have larger movements of Cp with yaw in these designs. Cp crossing the steering axis means that the wheel behaves one way in certain conditions and the opposite way in other conditions, and that's even worse than just having Cp on the wrong side of the axis.. One way around this would be to have the Cp start well ahead of steering axis and then come back with yaw..this would give you proportionally higher torque values at low yaw, but at least the handling would be predictable..


Here's a Cp plot we did for the 808 with Intelligent light...sorry, I'm limited to public domain stuff, but we had hundreds of these plots and graphs showing different rim shapes and the movements of the Cp's...it's super interesting stuff! Anyway, you can see that we designed for a Cp that was as low as possible and as close to the steering axis as possible without crossing it. The Cp moving rearward with yaw means that as side force increases the lever arm relative to steering axis is decreasing.. still looking for the graphs on this, but the goal is a graph that has pretty flat steering torque regardless of wind angle.

Long story short, the 808 in the Hambini test is now an ~8 year old shape...I have no doubt that there are faster shapes and designs at this depth. Heck, we had identified dozens of them before I left Zipp at the end of 2013.. but we never found one that was worth the tradeoff in handling, which is not in any way meant to say that it isn't possible. We were also beginning the study of the waveform inner diameter rims when I left and that more than tripled the design elements in a rim opening even more options for tuning. On a separate pathway we had begun using a mathematical technique called simulated annealing to seek other possible solutions for this when I left, which showed that for a given optimization parameter, there were groupings of solution sets that were competitive, but at the same time, the solution pathway we were one was pretty robust and hadn't been fully explored. Sadly you need infinite resources for that!!

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [hambini] [ In reply to ]
Quote | Reply
hambini wrote:
If we (and we have) go out and do several 40 second blasts and plot them together the correlation is no where near 0.97 against a power meter using your equations.
Does this help? I just created a quick Monte Carlo simulation in Excel to see what sort of standard deviation produces that outcome:



So that is with a range of true power values and number of data points similar to those in the study. R^2 = 0.97 is achieved with a standard deviation of measured power around 15W. Because it's a Monte Carlo simulation, the result changes each time I run it, so I've just run it 10 times and taken the average R^2, and it was 0.9751. So if anything, the SD should be a little higher than 15W. Does that make the amount of variation seem less implausibly low?
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Steve Irwin] [ In reply to ]
Quote | Reply
You could have just calculated the SD of all of the power values using the R^2 and SEE given in the paper.

That said, I have no clue what you are trying to achieve.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
That said, I have no clue what you are trying to achieve.
If I go out and aero test, then I can't really relate to an R^2 value, because I have a single true CdA, and lots of observations of measured CdA, so R^2 isn't relevant. What I can relate to is the standard deviation, because I can look at the standard deviation of my measured CdAs.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Steve Irwin] [ In reply to ]
Quote | Reply
What you have calculated does not tell you what you want to know.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
What you have calculated does not tell you what you want to know.
You'll have to elaborate, as I believe it tells me exactly what I wanted to know.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Steve Irwin] [ In reply to ]
Quote | Reply
It appears that you wish to know the between-run variation in power within a single individual, so you can relate that figure to the variability you see in field testing. The data in the paper, however, represents the between-run variation in power across multiple individuals. Furthermore, each individual only did a single run under a given condition (i.e., speed and direction).

Now if it is the within-run variation in power (either within or across individuals) that interests you, you can't determine that from the data either, as only the average value was used in further calculations.
Last edited by: Andrew Coggan: Sep 7, 18 11:44
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
You could have just calculated the SD of all of the power values using the R^2 and SEE given in the paper.

Or just read the paper:

"Mean values were 172.8 ± 14.7 W for SRM versus 172.0 ± 15.2 W for the model"

(I forgot that the overall means and SDs were in there.)
Last edited by: Andrew Coggan: Sep 7, 18 11:42
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
It appears that you wish to know the between-run variation in power within a single individual, so you can relate that figure to the variability you see in field testing. The data in the paper, however, represents the between-run variation in power across multiple individuals. Furthermore, each individual only did a single run under a given condition (i.e., speed and direction).

Now if it is the within-run variation in power (either within or across individuals) that interests you, you can't determine that from the data either, as only the average value was used in further calculations.
That is handled okay by what I did. Each random number is a sample from a normal distribution with a mean of the true power in the column to the left, and a standard deviation of the value in that cell. It's a single sample from the distribution that would be expected for that one person, i.e. the model is of variation within a single individual. The graph then calculates the r^2 across individuals like you did. So the model shows the variation within each individual that gives rise to that r^2 across individuals.
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Steve Irwin] [ In reply to ]
Quote | Reply
Picture two cyclists tested at multiple speeds, each under perfectly still conditions and while holding perfectly constant power on perfectly level ground. Also assume that the model is perfect, i.e., within a given individual the R^2 between modeled and predicted power is 1.00. Finally, assume that one cyclist is much heavier and less aerodynamic than the other. In this scenario:

1) the within-run variation in power is zero (assumed);

2) the between-run variation in power at a single speed is zero (corollary from first assumptions in first sentence);

3) R^2 between modeled and predicted power when you pool all of the data is less than 1.00.

IOW, as I said before the answer that you are seeking cannot be determined from the regression analysis, because the latter was performed by pooling averaged data across individuals.
Last edited by: Andrew Coggan: Sep 7, 18 12:53
Quote Reply
Re: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [Andrew Coggan] [ In reply to ]
Quote | Reply
Andrew Coggan wrote:
Picture two cyclists tested at multiple speeds, each under perfectly still conditions and while holding perfectly constant power on perfectly level ground. Also assume that the model is perfect, i.e., within a given individual the R^2 between modeled and predicted power is 1.00. Finally, assume that one cyclist is much heavier and less aerodynamic than the other. In this scenario:

1) the within-run variation in power is zero (assumed);

2) the between-run variation in power at a single speed is zero (corollary from first assumptions in first sentence);

3) R^2 between modeled and predicted power when you pool all of the data is less than 1.00.
Do you mean measured (the graph from the study is labelled as measured power vs predicted power)? In any case, I don't see how 3 follows. If the two are perfectly correlated, you'll just have a load of points (x1,x1), (x2,x2) etc, even if they've come from different riders, and r^2 will be 1.00.

Or are you saying each point on the graph represents some sort of average? Even if that were the case, averaging (x1,x1), (x2,x2) etc will still give (y1,y1), i.e. it won't move the correlation away from 1.00.
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [joshatsilca] [ In reply to ]
Quote | Reply
Thanks for all the insight Josh. Interesting point about "once we had a design that was nearly as fast as a disc". I'd never thought about it that way but, once you pull watts to spin out, many modern wheels are nearly as fast as a disc at low yaw so there really is not a lot left on the table in terms of aerodynamic gains.

When you say "wave form inner diameter rims" I assume you mean the 454 and 858 rim shapes? What are your thoughts on those wheels out of curiosity? How do you think that design fares with regards to managing the center of pressure?

I'm also curious to hear your thoughts on the new Bontrager XXX wheels as they make some big claims with regards to side force to drag but... as you pointed out... it's not just about side force but also where the center of pressure is and how stable its location is (something I'd literally never thought of until you just brought it up).

One more question, just because I'm curious to know if you and/or Zipp looked into it, what are your thoughts about wheel design and frame and fork interactions? Can they be done separately or should they be viewed as a system? Can wheel design be frame/fork agnostic? Are wider forks less sensitive to wheel selection?
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [joshatsilca] [ In reply to ]
Quote | Reply
joshatsilca wrote:
Long story short, the 808 in the Hambini test is now an ~8 year old shape...I have no doubt that there are faster shapes and designs at this depth. Heck, we had identified dozens of them before I left Zipp at the end of 2013.. but we never found one that was worth the tradeoff in handling, which is not in any way meant to say that it isn't possible.

Thanks for that. You probably have us much design and wind tunnel experience with aero wheels as anyone on the planet so it's great to have your input!

What are your thoughts regarding the premise of Hambini's testing protocol; that small oscillations in the flowfield and transient yaw movements (more closely modeling outdoor riding conditions) can have a significant effect on the drag and stability results?
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [rruff] [ In reply to ]
Quote | Reply
rruff wrote:
joshatsilca wrote:
Long story short, the 808 in the Hambini test is now an ~8 year old shape...I have no doubt that there are faster shapes and designs at this depth. Heck, we had identified dozens of them before I left Zipp at the end of 2013.. but we never found one that was worth the tradeoff in handling, which is not in any way meant to say that it isn't possible.


Thanks for that. You probably have us much design and wind tunnel experience with aero wheels as anyone on the planet so it's great to have your input!

What are your thoughts regarding the premise of Hambini's testing protocol; that small oscillations in the flowfield and transient yaw movements (more closely modeling outdoor riding conditions) can have a significant effect on the drag and stability results?


Good question. Despite Josh's protestations about leaving out Zipp, zipp.com is unfortunately a veritable desert of tech information that goes beyond marketing pablum.

There is a slight hint as to their thinking on one of their "whitepapers", :




Sort of interesting that they call the "normal conditions" 5-16.25 degrees. That's more consistent with Hambini than the ST-approved notion that 0-10 is "normal conditions." Especially that 0-5 is completely left out.

But I could be reading way too much into that cartoonish graph.
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [trail] [ In reply to ]
Quote | Reply
Zipp has always favored that high yaw regime. So did all the other wheel manufacturers AFAIK... until recently. I don't know if there was ever a scientific justification for it, but deep rims have very low drag in the 10-15deg yaw zone, so....

Finally some people actually went out and measured it: https://www.slowtwitch.com/...Yaw_Angles_5844.html

Paging Skippykitten (James Webb)! He's been developing his RiF aerostick for awhile now so he should have a very good idea regarding real wind conditions and variability.


Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [rruff] [ In reply to ]
Quote | Reply
rruff wrote:

Finally some people actually went out and measured it: https://www.slowtwitch.com/...Yaw_Angles_5844.html

Yes - that's what I meant by the ST-approved notion of yaw angles.

I just wish that page had more information. Few go much deeper than "bicycle mounted sensor". Only Flo, as far as I can tell, goes so far as to specify a specific model and sampling frequency. Flo calls 1Hz "high frequency," but that's *not* high frequency for transient analysis.

My point being that a vaned yaw/beta sensor already has some built in "mechanical low pass filtering". It's, by its nature, intended to be a steady-state direction sensor. Then for sampling rate it's not clear if data is being integrated over that one second, and 1Hz is just the recorded sampling rate, or if it's truly an instantaneous sample. If integrated, that's another filtering effect. In the Coggan paper the wind direction sensor was on a control tower somewhere, apparently?

My point here is I'm just wondering if Zipp knows more about this transient, more chaotic "wind buffeting" right at the surface layer. Suggested by Hambini. Rather than this smoothed vane-calculated vector, possibly integrated over a time period of at least a second.

We're jumping all over Hambini for not telling us much about his data collection apparatus and methods. But we're also, as far as I can tell, not eating our own dog food. No one has really specific apparatus or methods with sufficient detail to attempt complete replication of results. Except maybe Flo.
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [joshatsilca] [ In reply to ]
Quote | Reply
Hello joshsilca and All,

Future stall and gust control gains may be had by the automatic articulating of portions of the wheel (or vanes on spokes) to match changing flow conditions.

Did you experiment with this concept at Zipp?



While this slat design shown here was preceeded in prop planes .... the version here in the Me 262 ... was later copied almost exactly for the F-86 and FJ-2 and performance was excellent. (Personal experience USMC FJ-2)



The slats require no power other than the motion of aircraft and no control inputs from pilots ,,,, the relative simplicity of using gravity and reduced air pressure to deploy the slats and increased air pressure to retract it is a good example of using changing enviromental conditions to cause 'automatic' device changes to adapt for optimum performance.

For aircraft slats provide a higher angle of attack for wing sections and a delayed stall.

For bicycle wheels slats could be used to delay wheel stall in gusts and also have a dumping feature to prevent an upset with forceful gusts values beyond the stall controlled values.

Aircraft slats shown here have enough mass to move relatively slowly. Bicycle wheel slats could be light as a feather and move very quickly.

The slat has a counterpart found in the wings of some birds, the alula, a feather or group of feathers which the bird can extend under control of its "thumb". [Wiki]



A bicycle front wheel with both the characteristics of a full disk like a Zipp Super-9 Disk (low drag) and a Zipp 202 (good control in windy conditions) should sell well.

Cheers, Neal

+1 mph Faster
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [GreenPlease] [ In reply to ]
Quote | Reply
GreenPlease wrote:
One more question, just because I'm curious to know if you and/or Zipp looked into it, what are your thoughts about wheel design and frame and fork interactions? Can they be done separately or should they be viewed as a system? Can wheel design be frame/fork agnostic? Are wider forks less sensitive to wheel selection?

Great question! 😉

Scott
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [trail] [ In reply to ]
Quote | Reply
trail wrote:

Flo calls 1Hz "high frequency," but that's *not* high frequency for transient analysis.

My point being that a vaned yaw/beta sensor already has some built in "mechanical low pass filtering". It's, by its nature, intended to be a steady-state direction sensor. Then for sampling rate it's not clear if data is being integrated over that one second, and 1Hz is just the recorded sampling rate, or if it's truly an instantaneous sample. If integrated, that's another filtering effect. In the Coggan paper the wind direction sensor was on a control tower somewhere, apparently?

My point here is I'm just wondering if Zipp knows more about this transient, more chaotic "wind buffeting" right at the surface layer. Suggested by Hambini. Rather than this smoothed vane-calculated vector, possibly integrated over a time period of at least a second.

We're jumping all over Hambini for not telling us much about his data collection apparatus and methods. But we're also, as far as I can tell, not eating our own dog food. No one has really specific apparatus or methods with sufficient detail to attempt complete replication of results. Except maybe Flo.


Good point about how that sensor may bias low in its readings.
Last edited by: GreenPlease: Sep 7, 18 16:39
Quote Reply
Re: Updated: A Comprehensive (But Controversial) Wind Tunnel Wheel Shootout [trail] [ In reply to ]
Quote | Reply
trail wrote:
We're jumping all over Hambini for not telling us much about his data collection apparatus and methods. But we're also, as far as I can tell, not eating our own dog food. No one has really specific apparatus or methods with sufficient detail to attempt complete replication of results. Except maybe Flo.

I've got an Alphamantis Aerostick I'm dusting off for some "playing around". It's a pitot tube device, so shouldn't have any mechanical "low pass" action...IIRC, wind speed and yaw angle are transmitted at 4Hz on ANT+.

I'll have to check what the internal sampling frequency is and if the 4Hz transmitted data is averaged or downsampled.

I'll let everyone know what I observe...

http://bikeblather.blogspot.com/
Quote Reply

Prev Next