Hi Sitsofe, Indeed it is normal output. Even when using eyeballs only, if the blocksize is large, a few hundred IOPs of rounding makes for some confusing differences. Agreed it would be helpful to add a note in the docs. Thanks. Regards, Jeff -----Original Message----- From: Sitsofe Wheeler [mailto:sitsofe@xxxxxxxxx] Sent: Friday, July 21, 2017 6:36 PM To: Jeff Furlong <jeff.furlong@xxxxxxx> Cc: fio@xxxxxxxxxxxxxxx Subject: Re: Number of Significant Figures Hi Jeff, On 22 July 2017 at 00:09, Jeff Furlong <jeff.furlong@xxxxxxx> wrote: > > In commit d694a6a7c02f577b2bb5d0ad24331b775acf6869 the reported IOPs changed from 6 significant figures to 4 significant figures. So a fio summary file may change from 12345 IOPs to 12.3k IOPs. The attached patch will restore the original 6 significant figures (or maxlen in num2str.c). Which output is this in relation to - normal, terse or json? Note that if you're referring to normal then I'm afraid that output format is not considered stable and can arbitrarily change (the current change was made for compactness). The view is regular output is meant for eyeballs and users should use the json (or less flexible terse) output for machine parsing purposes. Now I look perhaps this needs to be made more explicit in the documentation... -- Sitsofe | http://sucs.org/~sits/ ��.n��������+%������w��{.n�������^n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�
