So perhaps I should add that the main thing the block sizes do is 
determine how non-contiguous the data is which needs to be read from disk 
in order to do the contouring.  The smaller the block sizes the more 
blocks the code needs to jump around disk to read, and that is slow.  The 
bigger the block sizes the more data that has to be read that isn't 
necessarily ever needed by the program.  It's a trade-off.  Quite possibly 
the UCSF output from bruk2ucsf typically achieves the better end of that 
trade-off.

Wayne

On Mon, 28 May 2012, Phelan, Marie wrote:

> Hi Markus
>
> I use ccpn analysis with ucsf format spectra. I don't do anything special, I just process my spectra using topspin and then convert the resultant 3rrr to ucsf using the standard bruk2ucsf program.
>
> As with most things there is an element of trial and error, I guess as Wayne says it depends on the spectra you are working with in the first place. In my experience the ucsf format works better than bruker for 'standard' 3d spectra associated with protein structure determination.
>
> Best wishes
>
> Marie
>
> NMR Centre
> Biological Sciences
> University of Liverpool
> Crown Street
> Liverpool
> L697ZB
> 0151 795 4398
> ________________________________________
> From: CcpNmr software mailing list [[log in to unmask]] on behalf of Markus Heller [[log in to unmask]]
> Sent: Monday, May 28, 2012 3:32 PM
> To: [log in to unmask]
> Subject: Re: ucsf or 3rrr
>
> Hi Marie,
>
> Could you please explain what works better?  That would be great...
>
> Thanks
> Markus
>
> --
> Markus Heller, Ph.D.
> Scientist, NMR
>
> Centre for Drug Research and Development (CDRD)
> 364--2259 Lower Mall
> Vancouver, B.C.
> Canada V6T 1Z4
>
> p: 604 221 7750 ext 122
> f: 604 221 7753
>
> www.cdrd.ca
>
> On 2012-05-28, at 1:23 AM, "Phelan, Marie" <[log in to unmask]> wrote:
>
>> Hi Markus
>>
>> For some unknown reason I find ucsf works a lot better with analysis than 3rrr format, so if you've got a choice that's what I would work with.
>>
>> Best wishes
>>
>> Marie
>>
>> NMR Centre
>> Biological Sciences
>> University of Liverpool
>> Crown Street
>> Liverpool
>> L697ZB
>> 0151 795 4398
>> ________________________________________
>> From: CcpNmr software mailing list [[log in to unmask]] on behalf of Wayne Boucher [[log in to unmask]]
>> Sent: Monday, May 28, 2012 8:00 AM
>> To: [log in to unmask]
>> Subject: Re: ucsf or 3rrr
>>
>> Hello,
>>
>> They are both blocked data formats which means they both ought to be
>> equally ok for Analysis.  (And I've never converted one to the other but I
>> imagine whatever the conversion process is used is unlikely to change the
>> actual block size.)
>>
>> Regards, Wayne
>>
>> On Mon, 28 May 2012, Markus Heller wrote:
>>
>>> Hello all,
>>> Sparky user here, but I have decided to give CcpNmr Analysis a try for my next assignment
>>> project, since I'm pretty impressed with its capabilities.
>>>
>>> What file format for the NMR data would you recommend and why?  My top two options are Bruker's
>>> 3rrr and sparky's ucsf.  I've pretty much never used nmrpipe, which eliminates a bunch of
>>> options.
>>>
>>> So, if you had to decided between 3rrr and ucsf, what would you pick and why?
>>>
>>> Thanks and Cheers
>>> Markus
>>>
>>>