Radiance (IES) vs. AGI

[bobby]

Has Anyone done a simulation comparing the result in AGI and Radiance?
I did for one specific room with similar reflectance for walls, roof, and ceiling and VLT for glazing both in AGI and IES. However, the results are different.......

[vajindarladdad]

Hi ,
I think you have sent a query to our support department as well.
This answer is for you & other users.

If you are doing a comparison study, be very careful that you maintain consistency between the 2 models. You should check the following are the same in each program.

1. Site orientation
2. Project location
3. Sky type
4. View taken (illuminance in the plan view will be looking at the floor in radiance, not the working plane. (use illuminance WP if necessary)
5. The surface properties, especially the glazing transmittance value
6. I am not familiar with AGI so I’m not sure if there are different levels of complexity you can use – radiosity to ray trace. Radiance uses ray tracing.
7. You should also check that the units are consistent. In radiance, even when in IP units the default units will be in lux so change this to fc in the menu of the image under units. This could be an easy explanation for a 10x difference.

Also , it depends upon input parameters from AGI.


I hope this has been of some help.

[bobby]

Thanks for your reply... But, my concern was has anyone done this comparison before? the items which you mentioned are the same in both softwares....

Thanks,

Bobby

[KasperS]

We use this program quite often at my university, and every time people get the same result. IESVE Radiance gives a way lower illuminance inside the rooms compared to other software. I have also compared it to different standards provided by some of the building institues in Denmark, and they also show that IESVE is way below.

I spoke to one of my professors once, and he said there was some talk about IESVE uses a very old version of Radiance, and that might be the problem.

But as you might have seen, if you bring it up to IESVE that it is not working correctly, you get the answer back "You aren't doing it correct, you should get some training". The support sent me back an analysis together with it. I tried to compare his results with the guide from SBi.
His model: at floor level.
Sbi: at 85 cm height.
Sbi and his model had almost the excact same results, which really shows that IESVE is not good enough.

[PCully]

Hi,

It is an older version of radiance that is integrated to VE but we're looking to update this quite soon, maybe you'll see a better result when that is available.

However, we have done tests on the results from RadianceIES and always been satisfied with the results. Sorry if you are not happy with the responses you've had from Support but very often with these queries we do find that the user is not comparing like for like and perhaps isn't aware of all the inputs and assumptions (again i apologise that this is an area IES are lacking that we don't make it easier to obtain reference and input data all the time)

Phil

[KasperS]

Hello

When do you expect the new version to be integrated? Is there a timeschedule or is it too early to say? I have to calculate the DF for an entire hospital, so it would make it a lot easier if i could use IESVE instead of another software.

From Kasper

[PCully]

I'd like to be able to help but it will be Spring next year at the earliest, there's a bit of work to do to get the latest version integrated and we'll want to do a fair bit of testing ourselves before it's released. Problem from my perspective is our decision makers will see w already have a Radiance product out there so it's a lower priority to update it - we'll keep pushing for it though.

If I get any update or clearer (confirmed) timescales I'll let you know.

Phil

[KasperS]

Thanks

[ZapBran]

Just for general interest (I'm not a lighting guy) I had a little google and found the Radiance release notes:
http://radsite.lbl.gov/radiance/refer/N ... Notes.html

I think Phil mentions above that IES use an older Radiance. One of my mates told me, so this is only hearsay and not Gospel, that IES use Radiance version 3.4 or 3.5 so they are a good few years behind any bug fixes and enhancements.

This might be a ridiculous suggestion but Flucs program might be an option although I have found that it is horrendously slow (I mean like a glacier slow) when the quality is higher.

[KasperS]

Well atm we don't have the Flucs at my work (i am the only one using IESVE, so sometimes it pretty hard to convince my boss to keep buying new ones), and i havent really worked with it before. But i am to do daylight calculations for 50-60 different rooms, but if Flucs can do it for an entire model at the time (or i can quee them) it could be used.

Phil, do you know if Flucs would give me a more accurate estimate of the daylight?

Kasper