Jmmlc said:
Or in equivalent wavelengths to evaluate the magnitude of the corresponding summing error
btw: Now I know that my PA midrange horns exhibit more than one wavelength of group delay at 230Hz, this explains all the summing problems and why adding a 2nd order allpass around 230Hz with a Q around 2.2 to the bass horn improves things a lot... (This group delay is spatial loading dependent too)
Attachments
Hello Eva,
It is a good idea but a bit difficult to implement as at high frequency there will be in many case a lot of noise on the group delay curve.
Best regards from Paris, France
Jean-Michel Le Cléac'h
It is a good idea but a bit difficult to implement as at high frequency there will be in many case a lot of noise on the group delay curve.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Eva said:
Or in equivalent wavelengths to evaluate the magnitude of the corresponding summing error
Jmmlc said:
Eva said:
Hi Jean-Michel and Eva,
Group delay expressed in units of time is good enough for me
.The raw data can of course be exported and manipulated external to Hornresp, if so required.
Kind regards,
David
PS - Thinking about it a bit more, I guess I could at least add the the equivalent distance value, and possibly even the number of wavelengths, to the group delay chart Sample tool results
.
David McBean said:
Greets!
I can understand why, I had to use the emergency dial-up for a bit and the forum page was taking 'forever' to load, change to 'post', etc., so waited till I had DSL again before trying to post this.............
Anyway, after ~4 min. it 'completed' at only 646 kb and got this error message when I tried to open it:
GM
PS - Used HR some more and still no virtual memory, etc. problems (he says with fingers crossed).
Attachments
Hi GM,
Many thanks for this. You have helped to confirm that there is definitely a problem downloading Hornresp from the new website via a dial-up connection. I will go back to my ISP and see what they can do about it.
Excellent!
I think it is probably unlikely that the two problems are related, but I have been wrong before.
Kind regards,
David
GM said:
Many thanks for this. You have helped to confirm that there is definitely a problem downloading Hornresp from the new website via a dial-up connection. I will go back to my ISP and see what they can do about it.
GM said:
Excellent!
GM said:
I think it is probably unlikely that the two problems are related, but I have been wrong before
.Kind regards,
David
Hy david,
I've been investigating the download-Issue for a few days now and have asked people all over the world to check the behaviour...
It seems that the routing of the servers at bigblog.com.au is veeery slow and somehow messed up. All people in europe I´ve asked so far have very slow response even when just looking at the site and having VDSL Connections. This might explain the behaviour of dial-up connections being unable to do correct downloads, since the amount of data can't be the issue here, even a search in google.com is more data to transmit than the simple blog-page and the few kB from the hornresp-File aren´t the problem either....
When calling the page from europe, most of the times the routing stops somewhere at sidney:
The rather high response times up to 400ms might be the caus why connections brake up and a download is messed up. A download-manager being able to re-connect should fix the problem, but this website seems to be veeeery slow, no matter what... Since most DSL/cable stuff has different or Up-to-date routing, caching and proxy-stuff, this might explain why a few people with dial-up connections had problems downloading...
to compare, this is google:
A friend of mine even had 1044ms repsonse time on the connection hong-kong australia...
I guess since this blog-site is a free offer to customers of bigpond ISP, they just don´t give too much attention to the connection and the server-speed.... They mention a difference between dial-up and cable/DSL Connections on their website concering the costs... but this doesn´t have much to do with the difficulties downloading, I guess it´s more a timeout-issue caused by the very slow response-times and the intense routing over very slow lines...
Ah btw.... this is the way the data traveled from my PC to your blog
you could try this in windows, to see what happens if you call it up from australia:
open up a "DOS-Prompt" (command line interface..) and type:
tracert www.dmcbean.bigblog.com.au
or use this webinterface which uses a java-app, though:
http://www.mapulator.com/
I've been investigating the download-Issue for a few days now and have asked people all over the world to check the behaviour...
It seems that the routing of the servers at bigblog.com.au is veeery slow and somehow messed up. All people in europe I´ve asked so far have very slow response even when just looking at the site and having VDSL Connections. This might explain the behaviour of dial-up connections being unable to do correct downloads, since the amount of data can't be the issue here, even a search in google.com is more data to transmit than the simple blog-page and the few kB from the hornresp-File aren´t the problem either....
When calling the page from europe, most of the times the routing stops somewhere at sidney:
The rather high response times up to 400ms might be the caus why connections brake up and a download is messed up. A download-manager being able to re-connect should fix the problem, but this website seems to be veeeery slow, no matter what... Since most DSL/cable stuff has different or Up-to-date routing, caching and proxy-stuff, this might explain why a few people with dial-up connections had problems downloading...
to compare, this is google:
A friend of mine even had 1044ms repsonse time on the connection hong-kong australia...
I guess since this blog-site is a free offer to customers of bigpond ISP, they just don´t give too much attention to the connection and the server-speed.... They mention a difference between dial-up and cable/DSL Connections on their website concering the costs... but this doesn´t have much to do with the difficulties downloading, I guess it´s more a timeout-issue caused by the very slow response-times and the intense routing over very slow lines...
Ah btw.... this is the way the data traveled from my PC to your blog
An externally hosted image should be here but it was not working when we last tested it.
you could try this in windows, to see what happens if you call it up from australia:
open up a "DOS-Prompt" (command line interface..) and type:
tracert www.dmcbean.bigblog.com.au
or use this webinterface which uses a java-app, though:
http://www.mapulator.com/
HORNRESP VERSION 22.10
Hi Everyone,
Hornresp Version 22.10 has just been released. Changes are:
* Impulse response can now be calculated off-axis.
* Corrected rather than standard wrapped phase is shown on the phase response chart.
* Both corrected and wrapped phase are shown in the phase sample tool results.
* Equivalent distance travelled and equivalent wavelengths are shown in the group delay sample tool results.
* Both corrected and wrapped phase are exported.
Product Number 2210-090615 refers. Please download from my old website (at the address below).
Kind regards,
David
Hi Everyone,
Hornresp Version 22.10 has just been released. Changes are:
* Impulse response can now be calculated off-axis.
* Corrected rather than standard wrapped phase is shown on the phase response chart.
* Both corrected and wrapped phase are shown in the phase sample tool results.
* Equivalent distance travelled and equivalent wavelengths are shown in the group delay sample tool results.
* Both corrected and wrapped phase are exported.
Product Number 2210-090615 refers. Please download from my old website (at the address below).
Kind regards,
David
Re: HORNRESP VERSION 22.10
Hello David,
This new version is nearly perfect!
Here attached a very smooth phase curve for a Le Cléac'h Fc = 160Hz horn (T = 0)
Something that could be interesting to have in a future version of Hornresp should be a preset for a constant velocity diaphragm...
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello David,
This new version is nearly perfect!
Here attached a very smooth phase curve for a Le Cléac'h Fc = 160Hz horn (T = 0)
Something that could be interesting to have in a future version of Hornresp should be a preset for a constant velocity diaphragm...
Best regards from Paris, France
Jean-Michel Le Cléac'h
David McBean said:
Attachments
Re: Re: HORNRESP VERSION 22.10
Hello David,
I know you are working on a constant velocity throat model for Hornresp. Thanks for that.
Do you think it will be a lot more wotk to add a constant acceleration throat model? I'll be so curious to see if simulation done with such models on Le Cléac'h horns will be more consistent with measurement (e.g. response curves at different angles from axis)
Best regards from Paris, France
Jean-Michel Le Cléac'h
Hello David,
I know you are working on a constant velocity throat model for Hornresp. Thanks for that.
Do you think it will be a lot more wotk to add a constant acceleration throat model? I'll be so curious to see if simulation done with such models on Le Cléac'h horns will be more consistent with measurement (e.g. response curves at different angles from axis)
Best regards from Paris, France
Jean-Michel Le Cléac'h
Jmmlc said:
Cordraconis said:
Hi Cordraconis,
Only if it doesn't require too much work - I'll have a look and see what would be involved. The volume of each segment is of course already given on the main Schematic Diagram when the mouse pointer is moved over the segment. The total system volume is also shown (see attached).
Kind regards,
David
Attachments
Re: Re: Re: HORNRESP VERSION 22.10
Hi Jean-Michel,
I'm not sure how much work would be involved - I will investigate further (I have a nasty feeling though that things might start to get complicated). In the meantime, I will release the constant velocity diaphragm option separately as Version 22.20.
Kind regards,
David
Jmmlc said:
Hi Jean-Michel,
I'm not sure how much work would be involved - I will investigate further (I have a nasty feeling though that things might start to get complicated). In the meantime, I will release the constant velocity diaphragm option separately as Version 22.20.
Kind regards,
David