x-rayman.co.uk

Posted: **03 Apr 2009, 03:12**

johnewarren wrote:Yep shelxh is the way forwards.

Actually, if you need SHELXH you don't have the latest version of SHELXL. I recommend you update your software. The latest version of SHELXL has a larger array B than the old version of SHELXH (thus SHELXH no longer exists, officially) and some other things have been improved too.

Cheers,

Peter

Posted: **03 Apr 2009, 08:31**

Hi Peter, I have this version:

Code: Select all: ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + XL - CRYSTAL STRUCTURE REFINEMENT - SHELXTL Ver. 6.12 W95/98/NT/2000/ME + + Copyright(c) 2001 Bruker Analytical X-ray Solutions All Rights Reserved + + 07-7 started at 10:24:22 on 03-Apr-2009 + ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Which is the latest version?

By the way, SHELX is obviously only complicated to obtain? It is necessary to send an application form via post.

http://shelx.uni-ac.gwdg.de/SHELX/#Obtain%20SHELX

Posted: **03 Apr 2009, 09:05**

Peter you must have something special or know something that we don't I still need to use shelxh for large structures and I have both versions if I stick to the standard builds. You must be on the inside track on this one?

Code: Select all: +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + SHELXL-97 - CRYSTAL STRUCTURE REFINEMENT - UNIX VERSION + + Copyright(C) George M. Sheldrick 1993-7 Release 97-2 + + 3abs started at 10:00:25 on 3-Apr-2009 + +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Code: Select all: ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + SHELXH-97 - CRYSTAL STRUCTURE REFINEMENT - HUGE UNIX VERSION + + Copyright(C) George M. Sheldrick 1993-7 Release 97-2 + + 3abs started at 10:01:16 on 3-Apr-2009 + ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

But if I use the non-standard build of shelxl_ifc:
running 4 threads on 4 processors

Code: Select all: ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + SHELXH-97 - CRYSTAL STRUCTURE REFINEMENT - HUGE UNIX VERSION + + Copyright(C) George M. Sheldrick 1993-7 Release 97-2 + + 3abs started at 10:02:29 on 3-Apr-2009 + ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

it is correct I don't need shelxh as it is shelxh in a form. Although I thought that only UNIX/Linux people could use it as it uses the intel fortran compiler.

PS
Does George know about the new version :lol:

Shelx Website wrote:SHELXH - version of SHELXL for very large structures
SHELXH is a special version of SHELXL for the refinement of very large structures (with more than about 10000 unique atoms). The only difference between shelxh.f and shelxl.f is the first FORTRAN statement in which the array dimensions are specified by means of a PARAMETER statement. If even SHELXH is not large enough, you will need to change the dimensions of the arrays A and B as explained in the comments at the start of the source file. Large versions of SHELXS, SHELXPRO and SHELXA may be created in the same way, but it is rather unlikely that they will ever be required. Further details are provided by comments in the respective sources. SHELXL will print a suitable error message if it is necessary to increase the dimensions of the large arrays A or B.

A little care and fine-tuning may be required so that such large structures can be refined efficiently. If the computer does not have enough physical memory available, or if the 'maximum vector length' is set too large, SHELXH will run in disk exercising mode. This 'maximum vector length' refers to the number of reflections that are processed in one vector run, which may be smaller than the number in the input/output buffer. Some trial and error is needed to set the maximum allowed value so that the physical memory is fully exploited with a minimum of disk I/O for the virtual memory swap file. This number is set as the fourth parameter on the L.S. or CGLS instruction, and should be a multiple of 8; a good value to try for a 64MB computer is 64 (the third number on the L.S. or CGLS instruction is almost always zero). The array B is used as working space for these vectors (CGLS and L.S.) as well as for the least-squares matrix (L.S.). If the array B is not big enough, the program will use a smaller maximum vector run.

Posted: **03 Apr 2009, 10:48**

Hi John,

If course I know something you don't. As for the topic at hand, the latest version of SHELXL is from May of 2008. The Version number is 2008/4:

Code: Select all: ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + XL - CRYSTAL STRUCTURE REFINEMENT - SHELXTL Version 2008/4 + + Copyright(c) 2008 Bruker Analytical X-ray Solutions All Rights Reserved + + test started at 06:24:56 on 03-Apr-2009 + ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

I am using SHELXL and XL synonymously here, which is not entirely accurate, as there are indeed differences in the code (for example, as far as I know, the Flack-X parameter cannot be negative in SHELXL, but can be <0 in XL). SHELXL is the free version, which you can obtain from the author after submitting an application by mail, fax or, if you want, you can scan your completed form and send the JPG to George via e-mail. If you are a Bruker user, you can obtain the latest version of XL from Bruker AXS.

There still is the possibility that array B will be too small for your structure (proteins...) and you may have to compile your own build of SHELXL (not difficult). To my knowledge, however, George is no longer compiling the SHELXH flavor of his program, and the fact that he did not get around to update his homepage in almost two years is neither a big surprise -- if you know how George prioritizes -- nor proof of the continued existence of SHELXH.

The latest version of SHELXL not only comes with a very big array B, it also has a few bug fixes (for example the sum of occupancies in the CIF file for atoms in PART-1 that come very close to a special position is now correct). In addition there is now an official multi-processor version, called XLMP, which is, unfortunately, not working properly for twins.

Cheers,

Peter

Posted: **03 Apr 2009, 10:52**

Thats brilliant, a new version of shelxl I'll get on to downloading it from George's ftp site ASAP.

Cheers Peter

Posted: **03 Apr 2009, 12:54**

Hi Peter,

pmueller wrote:If course I know something you don't. As for the topic at hand, the latest version of SHELXL is from May of 2008.

But it's still termed 97? Will there be a complete new release in the future?

SHELXL is the free version, which you can obtain from the author after submitting an application by mail, fax or, if you want, you can scan your completed form and send the JPG to George via e-mail.

I will try..

Posted: **03 Apr 2009, 13:05**

Rudi wrote:But it's still termed 97? Will there be a complete new release in the future?

Hi Rudi,

Yes, it is still SHELXL-97. Nothing major has changed in the program since the historic SHELX workshop in Goettingen in 1997, marking the official release of SHELX-97. That was just a few months after I joined George's lab, so SHELX-97 is the only version I really know.

Will there be a major new version in the future?
Official answer: I don't know -- Only George knows.
Unofficial answer: Yes, there are plans and ideas for a new version with new features but I would not hold my breath -- this is still a few years out.

Cheers,

Peter

Posted: **03 Apr 2009, 13:11**

johnewarren wrote:Thats brilliant, a new version of shelxl I'll get on to downloading it from George's ftp site ASAP.

Hi John,

As long as you are downloading from George's server: XPREP, SHELXS, SADABS, cell_now and especially TWINABS (the last version from November 2008 has a number of substantial improvements -- much better than the previous version!) are also new. The most recent update I am aware of is SHELXD (aka XM), which was updated less than two weeks ago.

Cheers,

Peter

Posted: **07 Apr 2009, 08:51**

Hi Peter,

George Sheldrick has sent me the Login for his server

http://shelx.uni-ac.gwdg.de/shelx97/

But I wondered, because you mentioned the last version from 2008, but on the server reads
e.g. for win.tgz last modified at 17-Mar-2006 14:18.

And the SHELXH executable does still live in the folder.

Or am I on the wrong server? But I used the adress in the E-Mail letter.

Posted: **07 Apr 2009, 10:13**

Hi Rudi,

If you want to use SHLEXH, don't let me stop you.

Cheers,

Peter

Posted: **07 Apr 2009, 11:12**

No, that's not the problem.

Now I have tried the SHELXL version which I have obtained from the server:

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
+ SHELXL-97 - CRYSTAL STRUCTURE REFINEMENT - WINDOWS VERSION +
+ Copyright(C) George M. Sheldrick 1993-7 Release 97-2 +
+ his started at 13:05:13 on 07-Apr-2009 +
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

This obviously the same release (Release 97-2) as shown by John above but not the latest release from 2008 which you have shown, Peter.

So, my question is where can I find it? Or is it only distributed within SHELXTL? I noticed that your one was from SHELXTL.

PS: I could also not find the other files you have mentioned above (XPREP, SHELXS, SADABS, cell_now, TWINABS)? I guess because I'm not a user of Bruker instrumentation?

Posted: **07 Apr 2009, 12:33**

Hi Rudi,

XPREP, XP, SADABS, TWINABS and cell_now are proprietary programs of Bruker AXS and as such cost money. In addition, the latter three are specific to Bruker detectors (well, SADABS can be used with DENZO or HKL2000 output as well) and won't help you much if you don't use Bruker (or Nonius or Siemens) equipment.

Non sequitur: it is interesting how many people who don't have a license use XPREP and XP. As Co-Editor of Acta Crystallographica I have seen quite a few figures prepared with XP (if you know the program you can tell!) from authors who did not have a license. In the software section, those authors all claimed to have used some other package (most of them Ortep-III), but nobody ever complained when I pointed out this "mistake".

Back to the point: I assume that, if you write an e-mail to the author asking for the 2008/4 version of SHELXL, mentioning that you are not a Bruker user, you may get an e-mail explaining how you can get your hands on the executable. Don't you have direct access to the author's younger brother and former grad student? Maybe he can help...

Cheers,

Peter

Posted: **07 Apr 2009, 13:16**

pmueller wrote:Non sequitur: it is interesting how many people who don't have a license use XPREP and XP. As Co-Editor of Acta Crystallographica I have seen quite a few figures prepared with XP (if you know the program you can tell!) from authors who did not have a license.

Yes, the figures from XP are specific.

Back to the point: I assume that, if you write an e-mail to the author asking for the 2008/4 version of SHELXL, mentioning that you are not a Bruker user, you may get an e-mail explaining how you can get your hands on the executable.

Hm, but we have still Siemens P4 four-circle diffractometers at our department. May be therefor we can update our SHELTXL package?

However, may be the 2008/4 version was not yet released for free but exclusively in the SHELXTL package?

Don't you have direct access to the author's younger brother and former grad student? Maybe he help...

Yes, that's right. I will ask him what to do.

Posted: **11 Mar 2010, 10:00**

Hi,
I am refining a large small molecule (C92H154N3Ni12O54) with many disordered groups. The trouble is that with my computer, I can only use xh in shlextl to refine it since it seems to be too bing for xL. Now, my question is, how good is xh? and how should I report this in the .cif file?
Is it ok to use XH to refine a small molecule?
Thanx in adavnce for the help!

Posted: **11 Mar 2010, 10:09**

ShelXL and ShelXH go back to the time when you set the size of your array hardcoded into your source. Therefore if you had a small amount of memory and set up a large array within it then you'd occupy a lot of memory space but with no actual data. So to get around this you could do two things:

Recompile your code with an array appropriate to the size of your structure each time you wanted to refine a structure above a certain size.
Setup two programs one with an average size array and the other with a massive array

It appears that George provided both options by releasing the source to allow option 1 and by creating shelXH for option 2.

So in short to the best of my knowledge there is no fundamental difference in how the refinement is undertaken, unlike CGLS/L.S. but there is a memory cost in using the ShelXH vs the ShelXL. One which was important in the old days and which could effect the processing speed of your refinement but which on modern PCs is probably not a problem at all.

If you use Linux you can use the ShelX_IFC code which allows you to parallel process your structure for super fast refinements. I've also made a ShelX_Cuda version which uses your graphics card (NVIDIA) to do some of the heavy lifting (more cores than your CPU).

x-rayman.co.uk

ShelxH - ShelxL - Shelx Versions

ShelxH - ShelxL - Shelx Versions

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

Re: Problem with the ACTA command

using xh to refine a large small molecule

Re: using xh to refine a large small molecule