| And there are many native concerns about the way the driver behaves;
For instance the rme driver clicked when it had dropouts; and always stayed phase accurate. lynx didn't fix that functionality for years.
There are also a couple other things in some of the benchmarks that came
out that require questions to be asked.
Those latency questions are assuming a perfect blank project, no disk, etc.
There is no reason that the second you add 1 or 2 channels it does start
dropping out. I found the lynx cards to be way more sensitive about
the hardware they were in; maybe pcie will fix that. Hopefully it would
fix the issues with multiple cards.
(They seemed to be much more of a pci hog than the rme cards were;
so maybe extra bandwidth will magically fix the cards problems.)
(Lynx knows they were there; they even added debugging tools to the
end user mixer.)
There are also things that have not been answered, and that have been
documented to operate quite differently between the leaders drivers on the benchmarks.
#1. Lynx says the driver is multi-cpu capable; so the question has been posed
but not answered if the stability of no dropouts would be based on a aggregate
of the total sum of all the cpu's or just what happens if core 0/1 is 50%.
(Comes down to how each manufacture threads their driver.)
(So assuming multi-core is off, and you're just using 1 cpu; you fill it
50%; lets say the driver begins to drop samples. If that same cpu hits
50% in multi-core mode, does it drop samples.)
#2. The behavior of lynx and their dropout situation has changed and is supposed to change again; sure one isn't supposed to get dropouts; but their
hw seems to be far more prone to it. (am example is the big samplitude issue, but many people experienced it with cubase, runaway dropouts under no load, etc)
#3. routing just got added, so you could have more than 4 monitoring sources;
since it's pcie, is the card going to finally support a routing matrix like rme,
anything to anywhere, etc, at it's highest sample rate?
If I was going to buy a pcie card; I'd kind of want to feel that all the arcticture
issues with the aes16 were solved with the aes16-pcie; and I'm under
the impression none of them will be.
You know, hard issues to solve or not a 1-2 year turn around isn't acceptable;
I don't know how many times I was told they were to busy with aurora.
It's obviously their new cash cow; doesn't have the riddled issues the old
interface did; nor the support issues; I for the life of me don't see
why they even bother releasing another card, based on the market.
And well, I guess it hasn't happened yet either.
But I'll say this; they bug fix process takes as long as their announcement to delivery. |