Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[calvin]

So let me first give my comments to
For Pace; my idea of improvement is:
If I understand the meaning of "trigger point" correctly, it should be 95m instead of 995 m.
Maybe you just mistyped it... ?
Apart from that I think, your suggestion will result in a better pace calculation.
Because it is based on the idea to take a longer distance for calculation.
And for my personal point of view, the disadvantage of having an updated display only
every 100m is not a problem. For me it's fine.
I just would suggest to add just a little bit more intelligence to this method:
Is it not possible to include the real measured distance into the calculation ?
So instead of assuming a fixed value of 100m, to calculate with the real value of maybe
98m or 104m. It cannot be so difficult... And it would result in a perfect value !

For Pace1K; my idea of improvement is:
Also here, I'm not sure whether I understood correctly...
First I don't understand what you mean with "trigger point base on Auto Lap " ?
Does it mean you want to somehow combine the two functions "Auto Lap" and "Pace1k" ?
If yes, I cannot agree ! Please remember, that AutoLap may be configured in different modes
and (of course) different values.  But I think, I misunderstood...

But there is one more problem. When I follow your description, I understand that pace1k
will be updated only after a full km is reached. If this is true, I have to tell you:
This is not what we wanted to have ! This is then the same as AutoLap configured to 1km.

As a summary... 
When Pace1k is so complicated ( I do not believe !), maybe we should put it aside for a
while and have a more detailed look at the optimization of "normal" pace.
(Remember: Pace1k was suggested, because "normal" pace wasn't useable. So, if normal
pace becomes okay, maybe we can skip pace1k).
Perhaps you can think about my suggestions for "normal" pace (include real distance into
calculation) and give us a test version, so that we can see how it works.

Regards, Hubert

Hubert, you are right on the spot, I agree with you.

[Roger C]

Hi all !

Some days ago Geoffrey made a suggestion for pace and pace1k:

Hi, all

After my deep thinking and observation, I come out the solution for pace and pace1K:

There are the concerns affect judgement:
1. Memory size is almost full; no space to add complicated alogrithm
2. GPS accuracy
3. Signal strength
4. CPU power???

To eliminate the all concern and optimize calculation:
For Pace; my idea of improvement is:
time(1)=Time for first 100m (trigger point is 995m according to average GPS inaccuracy)
Pace(1)=time(1)*10 (00:00/km), update display;

time(2)=Time for 2nd 100m (trigger point is 995m according to average GPS inaccuracy)
Pace(2)=time(2)*10 (00:00/km), update display;
:
time(n)=Time for Nnd 100m (trigger point is 995m according to average GPS inaccuracy)
Pace(n)=time(n)*10 (00:00/km), update display;

For Pace1K; my idea of improvement is:
time(1)=time for first 1km (trigger point base on Auto Lap with cumulate error elimination
P1K(1)=time(1), update display;

time(2)=time for 2nd 1km (trigger point base on Auto Lap with cumulate error elimination
P1K(2)=time(2), update display;
:
time(n)=time for Nnd 1km (trigger point base on Auto Lap with cumulate error elimination
P1K(n)=time(n), update display;

Please let me know what about you think.

Sincerely,

Geoffrey 

And he asked for some comments...

So - please: He offers the chance to affect the next step for the functions of our watch !

Take the chance to give your comments ! 
Do you agree ? Is it that what you want to have ? Or do you have better ideas ?
Let us help Geoffrey and his team to find a good solution !
... for the benefit of all of us !

Regards, Hubert

Well, if we knew exactly how the pace and pace1k is calculated now, compared to this new 100m and 1000m update-intervall "improved version" then i would come with some ideas.

But like you and others said before, this pace1k thing came from the not so reliable pace calculation ,and made that mode useless.

So how is pace calculated now ? does it comes directly translated from the unstable speed calculation in the watch ? or are there any complicated algorithm involved ?

An improved pace mode that doesnt jump up and down so much, and is calculting from the real value that Hubert wrote, but with longer intervalls ,398m/106sec = 4,26 km/h , 404/110sec = 4.32 km/h (calculation made here: http://www.coolrunning.com/engine/4/4_1/96.shtml )(and updating on the watch every 3-4 secounds) , would be best to focus on, before a pace1k mode is implemented

Or isnt that almost what pace1k should be like ? with real value 998m/240 sec = 4 min/km and updating every 3-4 sec with a new real value.....

Algorithm like 1004m-2002m = 998m/240sec  and 4 sec later , 1020m-2028m 1008m/244sec and 4 sec later 1040m-2044m = 1004m/242sec and so on....

But for me the watch works great without this functions

Good night or good morning , see ya

Roger C

[snow_freak]

But for me the watch works great without this functions

Roger C,
I agree with you! For me its important to have an accurate lap pace with autolap 1k. The rest is done by "pace feeling".
Improvements of the actual pace would be nice, but before discussing the possibilities, we need to know how the current calculation works.

Regards, snow_freak

[TWG_TECH8]

I want to get the famous firmwareX.
Geoffey can you please send it to me?

send to:
luis.leite [at] gmail.com

Hello Calvin,

I sent, please check it.

Sincerely,

Geoffrey 

[Trinap21]

Hello, I'm a new gps user and I have the gs-625M and I use it for my running trainyng and trail walking. but I would like to know if the new firmware solve the problem's that I find. 1š when I use the back track, the watch send me rigth to the start point and not by the path that I took. 2š the graphic of the trainning gym software can show on the base km instead of the minutes?

Thank's

[TWG_TECH8]

Thanks for you all good suggestions!
@Roger:
It is a good tool, thanks!

Right now I come out the solution for pace:
1. Calculate the distance (d1) for first 3min (To get reliable value about distance)
2. Get pace=d1/3min;
3. Shift 3sec to get the distance (d2) for full 3min;
4. Get pace=d2/3min;
:
Pace(n)=dn/3min

Pace1K keep 1 full km to calculate.

Sincerely,

Geoffrey

[TWG_TECH8]

Hello, I'm a new gps user and I have the gs-625M and I use it for my running trainyng and trail walking. but I would like to know if the new firmware solve the problem's that I find. 1š when I use the back track, the watch send me rigth to the start point and not by the path that I took. 2š the graphic of the trainning gym software can show on the base km instead of the minutes?

Thank's

Hello,

Thanks for using our product.
Here is my answer to your question:
@1: Firmware X has been adding Smart Track Back alogirthm which can take you point by point to your destination!
@2: The current Software doesn't has distance instead of time in X axis of chart.
But we have plan to add this feature.

Hope my answer can help you.

Sincerely,

Geoffrey

[Trinap21]

[/quote]

Hello,

Thanks for using our product.
Here is my answer to your question:
@1: Firmware X has been adding Smart Track Back alogirthm which can take you point by point to your destination!
@2: The current Software doesn't has distance instead of time in X axis of chart.
But we have plan to add this feature.

Hope my answer can help you.

Sincerely,

Geoffrey
[/quote]

Hi, thank's, your answer was very helpfull. I'm looking forward for the new update of the training gym software, in the meanwile could you send me the new firmware X. Thank's

[Hubert]

Hi Geoffrey,

Right now I come out the solution for pace:
1. Calculate the distance (d1) for first 3min (To get reliable value about distance)
2. Get pace=d1/3min;
3. Shift 3sec to get the distance (d2) for full 3min;
4. Get pace=d2/3min;
:
Pace(n)=dn/3min

So, that means, we will have a value for pace, updated every 3 sec.
And it will show the average pace of the last 3min ?  Of course displayed as Km/Min. ?

From my point of view, this will improve the current solution. So it's okay to have it !

On the other hand...  (I'm sorry, there is always an "on the other hand"...)

... I think, that 3Min. might be too long for some situations:
For speed training I sometimes run short distances like 200m (< 50 sec.)
or 400m (< 100 sec.) and it would be nice to get a feeling of pace for those
distances too. And as you can easily see, here 3Min. are too long !

I know, that a longer distance makes the pace value more reliable.
But a shorter time applies more to the basic idea of current pace.
Basicly we want to see the pace of this moment, not of the last 3min.
So, we have to find a compromise of these two requirements.
And 3min. are not so bad... But what about 2Min. ? Or 1Min. ?

You know: My favorite solution would be a configurable value, so that users can
decide what they consider to be more important.

However - what ever you want to implement: It will be better than what we have now.

Pace1K keep 1 full km to calculate.

Sorry - but after the long and detailed discussion about this subject,
this statement was a little bit to short for me. 
I do not understand, what you want to do here 

Best regards, Hubert

[ultrapetita]

I agree with Hubert's comment above.

I think the "pace3min" is useful, but in addition to that, I would like to see some more up-to-date information on my running.

Thus, my proposal would be to have the pace3min as Geffrey proposed showing a stable "cruising" pace and another pace item, which would be improved (smoothed) from the current pace. (It can't be too complicated for the CPU to average 1-20 samples of currently displayed pace values - If there is no possibility to have moving average, this smoothed pace could be updated eg. once/10-20sec.)

Geoffrey, have you tried to smooth the currently displayed pace item? How do you calculate the pace/speed at the moment or does it come directly from NMEA information?

So, it seems that we come to the starting point of this discussion... how to get the pace value smoothed...

[snow_freak]

I think better than average and moving average for smoothing the current pace would be a median algorithm.
http://en.wikipedia.org/wiki/Median
It is very robust against spikes as they are caused by the deviation of the GPS signal. I dont know if the demand of CPU Power is too high for this alorithm.

[Longjog @ GS-Sport-Servicecenter]

Hello together,

the reason for pace1k was or is because the GPS-error is to big to get usable values for pace if you move in slow speeds like you do while running. So the values for pace are not useable at the moment.
I ask myself how Garmin is solving this problem or havenīt them solved the problem? Does anybody know if the Forerunner 205 / 305 or the 310XT show the actual pace while running in true values without jumping more than 10sec/min when running with a constant speed?

Regards

Longjog

[Hubert]

I ask myself how Garmin is solving this problem or havenīt them solved the problem? Does anybody know if the Forerunner 205 / 305 or the 310XT show the actual pace while running in true values without jumping more than 10sec/min when running with a constant speed?

Regards, Longjog

I also had this question in my mind and I will try to ask somebody. I know, that speed calculation in general is a problem
and that others try to solve it by an additional device (like foot pad). I don't like this solution because I don't want to carry
and maintain an extra device, so I forgot who uses this solution...

Regards, Hubert

[snow_freak]

I ask myself how Garmin is solving this problem or havenīt them solved the problem? Does anybody know if the Forerunner 205 / 305 or the 310XT show the actual pace while running in true values without jumping more than 10sec/min when running with a constant speed?

In Forums likes this: http://forum.runnersworld.de/forum/pulsuhren-gps-trainingssoftware/ the problem is also discussed. And it is not solved by Garmin!

[TWG_TECH8]

Hello,

Thanks for using our product.
Here is my answer to your question:
@1: Firmware X has been adding Smart Track Back alogirthm which can take you point by point to your destination!
@2: The current Software doesn't has distance instead of time in X axis of chart.
But we have plan to add this feature.

Hope my answer can help you.

Sincerely,

Geoffrey
[/quote]

Hi, thank's, your answer was very helpfull. I'm looking forward for the new update of the training gym software, in the meanwile could you send me the new firmware X. Thank's
[/quote]

Hi

Thanks. To satisfy customer's request is our responsibility and customer's encourage is our source to drive us progress. Now we are working on re-structre on software to reduce resource occupation and with more features. Hope it can be finished soon.
About firmware X, can you send me a mail and please write your email address in the mail because I will always get system respresentative address (likes Erick Madrid).

Sincerely,

Geoffrey