<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
	<channel>
		<title><![CDATA[RME User Forum — How to contact RME software developers?]]></title>
		<link>https://forum.rme-audio.de/viewtopic.php?id=36242</link>
		<atom:link href="https://forum.rme-audio.de/extern.php?action=feed&amp;tid=36242&amp;type=rss" rel="self" type="application/rss+xml" />
		<description><![CDATA[The most recent posts in How to contact RME software developers?.]]></description>
		<lastBuildDate>Tue, 08 Nov 2022 07:29:43 +0000</lastBuildDate>
		<generator>PunBB</generator>
		<item>
			<title><![CDATA[Re: How to contact RME software developers?]]></title>
			<link>https://forum.rme-audio.de/viewtopic.php?pid=193550#p193550</link>
			<description><![CDATA[<p>I have been contacted, thanks!</p>]]></description>
			<author><![CDATA[null@example.com (robpap)]]></author>
			<pubDate>Tue, 08 Nov 2022 07:29:43 +0000</pubDate>
			<guid>https://forum.rme-audio.de/viewtopic.php?pid=193550#p193550</guid>
		</item>
		<item>
			<title><![CDATA[How to contact RME software developers?]]></title>
			<link>https://forum.rme-audio.de/viewtopic.php?pid=193374#p193374</link>
			<description><![CDATA[<p>Does anyone know how to forward my message below to the developers of the ASIO driver hdsp32?</p><p>Thank you<br />-R</p><p>----</p><p>Feature request: memory-aligned buffers for ASIO driver (hdsp32.exe)</p><p>The support staff told me to post my message here for the RME developers:</p><p>When my C++ ASIO host is calling ASIOCreateBuffers(), I noticed that the Hammerfall DSP ASIO driver (for my HDPSe AES PCIe card) allocates buffers whose starting memory address is aligned to 16 bytes (the default).</p><p>It would be great to have them *aligned to 64 bytes* (the cacheline size) for the following reasons. </p><ul><li><p>First, it is backward compatible, as 64-alignment is also a 16-alignment, so ASIO hosts using the16-aligned buffers still work.</p></li><li><p>Second, cache performance improves on small buffers. For example, using 32 samples in the Hammerfall DSP ASIO driver, each channel has a buffer size of 128 bytes. Due to the fact the buffers are 16-aligned, these 128 bytes are hosted in *three* consecutive cachelines, whereas 64-aligned buffers require *two* cachelines. In other words, one third of cacheline loads is automatically saved by ASIO hosts in this way (no change is needed in their code).</p></li><li><p>Third, some specialized AVX2 instructions that require 32-byte alignment could be used, whereas this is not possible now.</p></li></ul><p>In summary, it would be very helpful if you could make this simple change in the next release of the Hammerfall DSP ASIO driver. Performance would have a benefit with no drawbacks (at least I could not see any of them so far).</p><p>Thank you<br />-R</p>]]></description>
			<author><![CDATA[null@example.com (robpap)]]></author>
			<pubDate>Wed, 02 Nov 2022 08:52:40 +0000</pubDate>
			<guid>https://forum.rme-audio.de/viewtopic.php?pid=193374#p193374</guid>
		</item>
	</channel>
</rss>
