<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@font-face
{font-family:Helvetica;
panose-1:0 0 0 0 0 0 0 0 0 0;}
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
span.EmailStyle20
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;
mso-ligatures:none;}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="blue" vlink="purple" style="word-wrap:break-word">
<div class="WordSection1">
<p class="MsoNormal">In that case, it should be easy to write up a few specific astronomical use cases. And collect citations to pertinent literature.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Does FITS need to wait on the new sub-16-bit standard before taking action?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Rob<o:p></o:p></p>
<div style="mso-element:para-border-div;border:none;border-bottom:solid windowtext 1.0pt;padding:0in 0in 1.0pt 0in">
<p class="MsoNormal" style="border:none;padding:0in"><o:p> </o:p></p>
</div>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<div id="mail-editor-reference-message-container">
<div>
<div>
<p class="MsoNormal" style="margin-left:.5in">On 7/24/25, 5:08 PM, "fitsbits" wrote:<o:p></o:p></p>
</div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
<div>
<div>
<p align="center" style="margin-left:.5in;text-align:center"><strong><span style="font-family:Helvetica;color:red">External Email</span></strong><b><span style="font-family:Helvetica"><o:p></o:p></span></b></p>
<p class="MsoNormal" style="margin-left:.5in"><span style="font-family:Helvetica"><o:p> </o:p></span></p>
<div class="MsoNormal" align="center" style="margin-left:.5in;text-align:center">
<span style="font-family:Helvetica">
<hr size="0" width="92%" align="center">
</span></div>
</div>
<p class="MsoNormal" style="margin-left:.5in">Malcom:<o:p></o:p></p>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">Just because we are old does not mean that we are not working on state-of-the art software. The reason for wanting 16-bit floats is for this specific reason. Radio astronomy software would benefit from 16-bit floats.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">All:<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in">As I noted previously, the IEEE is working on sub-16-bit (<= 15) floating point formats for AI. There is clearly a need for such data types in the machine learning community. Currently, hardware vendors are using
various sub-16-bit formats. The new standard is meant to standardize on a common one, just like IEEE 754 did decades ago.<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"> -- Paul<o:p></o:p></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
</div>
</div>
<p class="MsoNormal" style="margin-left:.5in"><o:p> </o:p></p>
<div>
<div>
<p class="MsoNormal" style="margin-left:.5in">On Thu, Jul 24, 2025 at 6:32 PM Malcolm J. Currie via fitsbits <<a href="mailto:fitsbits@listmgr.nrao.edu">fitsbits@listmgr.nrao.edu</a>> wrote:<o:p></o:p></p>
</div>
<blockquote style="border:none;border-left:solid #CCCCCC 1.0pt;padding:0in 0in 0in 6.0pt;margin-left:4.8pt;margin-right:0in">
<p class="MsoNormal" style="margin-left:.5in">Being (allegedly) retired and having little contact with the<br>
astronomical programming world post COVID, it's no surprise that<br>
I too was unaware of 16-bit floats before this discussion.<br>
<br>
My sympathies are with Walter's view, as I'm inclined towards regarding<br>
FITS as an interchange format rather than a processing format. My bias<br>
is partly due to having our own flexible and extensible format in<br>
Starlink. I'm wary of additions to the standard that have a quite<br>
limited usage solely for FITS as a working format.<br>
<br>
I agree with the points and questions made by Rob, especially how widely<br>
would these two additions be used and evidence of their benefits. A<br>
document for discussion covering these points, how the new data types<br>
would be implemented, and the implications for existing software<br>
packages should be presented. In many packages the float16 FITS data<br>
would presumably have to be converted to another data type, or say it's<br>
not supported.<br>
<br>
Starlink's data system is now built on HDF5 (rather than the in-house<br>
HDS that preceded it). A quick search turned up an RFC to add 16-bit<br>
floats to HDF5, and a long user discussion of the RFC, but I've yet to<br>
find HDF's justification for the introduction of float16, or whether it<br>
has been implemented.<br>
<br>
Wikipedia, that trusty source of knowledge and wisdom,<br>
(<a href="https://en.wikipedia.org/wiki/Half-precision_floating-point_format" target="_blank">https://en.wikipedia.org/wiki/Half-precision_floating-point_format</a>)<br>
relates that the concept has been around as early as 1982. Also I'd<br>
forgotten IEEE 754 had a binary16, presumably because there wasn't a<br>
perceived need and it wasn't in the floating-point addition to FITS.<br>
We already had BSCALE/BZERO to increase the dynamic range. The<br>
Wikipedia entry does give some reasons for using 16-bit floats, such as<br>
machine learning, and lists programming languages that support it, but<br>
I'd certainly like to hear more on potential benefits in astronomy.<br>
<br>
Lucio:<br>
> That is true, and it was what I referred to as the early times when one<br>
> could write a reader of FITS magtapes on 36-bit or 60-bit mainframes. Or<br>
> different endianness.<br>
<br>
Hence the strange, to a young audience, 2880-byte logical record length.<br>
That accompanies the restrictive 80-byte headers that must look so<br>
antiquated to postdocs and students. Are they doing their own thing<br>
rather than join in the FITS discussion? Most of the participants in<br>
this thread are of, let's say, a certain vintage.<br>
<br>
Lucio:<br>
> Also I seem to remember in the past there was a rule, or at least a<br>
> practice, that a new feature has tp be supported by two independent<br>
> existing implementattions.<br>
<br>
Indeed.<br>
<br>
>> And finally, finally, discussions like this and about JPEGXL demonstrate<br>
>> that we need to revive the IAU FITS WG.<br>
><br>
> And that's even more true than all the rest, specially for a delicate item<br>
> like basic data formats.<br>
<br>
Yes but who would serve on it? There needs to be a mix of decades of<br>
FITS experience, combined with a new generation who work with<br>
cutting-edge data and state-of the-art tools or who are designing<br>
upcoming major data-generating projects.<br>
<br>
Malcolm<br>
<br>
<br>
_______________________________________________<br>
fitsbits mailing list<br>
<a href="mailto:fitsbits@listmgr.nrao.edu" target="_blank">fitsbits@listmgr.nrao.edu</a><br>
<a href="https://listmgr.nrao.edu/mailman/listinfo/fitsbits" target="_blank">https://listmgr.nrao.edu/mailman/listinfo/fitsbits</a><o:p></o:p></p>
</blockquote>
</div>
</div>
</div>
</div>
</body>
</html>