Alright fair enough, sorry if I came off as a jerk
No problem
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
AuzenTech Prelude - Street Date?
- Thread starter MetalDwarf
- Start date
No problem
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
I wonder why some people would immediately dis something without even trying it first OMG my X-Fi gives me 30FPS more in games, now that I use a software encoder, I can only get 20FPS more in games Yeah right, so much for hardware acceleration....
OMG my X-Fi gives me 30FPS more in games, now that I use a software encoder, I can only get 20FPS more in games Yeah right, so much for hardware acceleration....What? "30 FPS" more than what?
Perhaps, as you own both cards, it might be a good idea to see about putting your money where your mouth is. Do some benchmarking with the X-Meridian alone, the X-Fi alone, then benchmark with the X-Fi pumping out into the X-Meridian whilst encoding. You'll inevitably need to resort to FRAPS runs for most games, but I'm positive there are a couple titles out there that can be timedemoed with audio, so that would obviously be a good place to start.
You're in a unique position here to attempt to set the record straight regarding hardware acceleration, as you're a typical gamer, with a typical rig, with both sound cards, but you only downplay hardware acceleration without ever providing any sort of data regarding its actual advantage. You also state quite matter-of-factly that there's no performance degradation or latency increase for software encoding, and you haven't justified that either.
Perhaps, as you own both cards, it might be a good idea to see about putting your money where your mouth is. Do some benchmarking with the X-Meridian alone, the X-Fi alone, then benchmark with the X-Fi pumping out into the X-Meridian whilst encoding. You'll inevitably need to resort to FRAPS runs for most games, but I'm positive there are a couple titles out there that can be timedemoed with audio, so that would obviously be a good place to start.
You're in a unique position here to attempt to set the record straight regarding hardware acceleration, as you're a typical gamer, with a typical rig, with both sound cards, but you only downplay hardware acceleration without ever providing any sort of data regarding its actual advantage. You also state quite matter-of-factly that there's no performance degradation or latency increase for software encoding, and you haven't justified that either.
MetalDwarf
[H]ard|Gawd
- Joined
- Jan 20, 2002
- Messages
- 1,072
I'm all for "ship it when it's DONE"
but im not such a big fan of
it will be out in MAY
no wait JUNE
no sorry let try JULY
just say Q2 2007 and when you KNOW you are going to be able to meet the a date then release the date.
this is like christmas being on Dec 25 and Santa does show up till March 17th.
but im not such a big fan of
it will be out in MAY
no wait JUNE
no sorry let try JULY
just say Q2 2007 and when you KNOW you are going to be able to meet the a date then release the date.
this is like christmas being on Dec 25 and Santa does show up till March 17th.
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
I'm using a 24" LCD with 1920x1200 resolution... Have you ever wondered why X-Fi reviewers normally use a low resolution to benchmark the card?
Lone_Bullet.be
Limp Gawd
- Joined
- Jul 26, 2005
- Messages
- 235
The May/June delay was because of the revision... and...i'm afraid i can't back it up about the drivers. I red it somewhere... anyway see it as speculation for now. My apologies.
The Tech Report did not. EDIT: Wait, yes they did...scratch that.Have you ever wondered why X-Fi reviewers normally use a low resolution to benchmark the card?
Are you willing or not willing to provide additional data?
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
I don't want to waste my time to do all the testing again with proper documentation. If you want to buy an X-Fi to get a better sound in games that support EAX >2.0, I don't have any objection with that eventhough most games I played don't even support EAX. However if you want to buy an X-Fi to get a FPS boost, my observation shows me that the real world performance is not that big and negligible when you use a high resolution because you will be limited by your GPU anyway. You would get a performance boost at a lower resolution but if you already have a powerful GPU, you would probably already get 80+ FPS even without X-Fi at low resolution. I really doubt that it would make any difference in games when you get 80+FPS without X-Fi and 90+FPS with X-Fi. If you don't already have a powerful GPU, I think that getting a better GPU will give you more performance boost than getting an X-Fi.
So, you've done testing already? Can you link us, or provide a summary of some sort?
I understand, but I don't think that's the issue here. Remember that this discussion evolved because you stated that the software encoder (one that requires calculations, operational memory and buffers) has no performance loss or latency increase, and I suggested that you do a few benchmarks for us to validate that, and you declined.
And that would be pretty much the only scenario you'll experience on your particular system, as your proc and graphics card aren't well-matched. As the GPU becomes the limiting factor, the CPU begins encountering a situation where it has enough free cycles to spend on somewhat frivolous things. If you had a more balanced system, like if you had an X1950, or maybe an 8800 GTS 320, you'd balance the equation, so to speak, and begin to see a somewhat wider performance spread with sound on and off. There are people running machines that are unbalanced at the other end of the spectrum -- pairing relatively slow CPUs with relatively fast video cards, and it's these people who will experience some potentially significant performance degradation with software sound processing. It depends entirely on how you define "significant", but it's certainly present enough to be measurable, even on the burliest rigs.
But, again, I don't think that's specifically what we're discussing here.
I wonder what kind of hit I get with my rig using software eax. I wish that I had an x-fi atm to test. I run 1920x1200 whnever the game supports it. 1600x1200 in 2142 all high settings seems to stay around 100fps but COD2 at 1920x1200 with all high can dip down to the 50s on some of the high end mod maps with lots of smoke. Heck, I can even get it down to the teens when I try to force 16x aa and af.
6400 @ 3ghz, 8800GTX 625/1000, 2 gigs @ 666 fsb for the OC, 74 gb raptor , XP PRO, Supersampling aa.
I have HIGH hopes for the prelude as I am holding $$ for it atm.
6400 @ 3ghz, 8800GTX 625/1000, 2 gigs @ 666 fsb for the OC, 74 gb raptor , XP PRO, Supersampling aa.
I have HIGH hopes for the prelude as I am holding $$ for it atm.
I have an X-Fi and tried it... It is not a driver, just a simple software, no performance loss nor high latency
Look, even a hardware encode is going to add latency and drop performance at least somewhat. How much exactly depends on a number of factors, but there is simply no such thing as an encode layer that is truly instantaneous.
I don't know you, but this combined with the fact that you don't even provide the name of this magical "space and time defying" software you are talking about, let alone a link, and just tell him to "search" tells me that you are an ordinary bullshitter.
Shakey_Jake33
n00b
- Joined
- May 19, 2007
- Messages
- 26
There will always be some sort of performance loss using software encoding... it uses CPU cycles, not voodoo magic.
The key will be exactly how much resources the encoding takes, while ultimately means we'll just have to wait and see, and have faith in Auzentech.
The key will be exactly how much resources the encoding takes, while ultimately means we'll just have to wait and see, and have faith in Auzentech.
And truth be told an AC3 encode isn't that processor intensive at all, relatively speaking. The biggest problem with the driverless solution that alg7_munif is suggesting is that without a driver, I don't see how you'd be able to get away without having to pass the audio through about 3 different buffers (dsound buffer, PCM > AC3 buffer, AC3 > spdif buffer.)
A driver on the other hand should be able to do all of that with just one audio buffer.
A little more detail: You can't convert audio from PCM to AC3 with just one sample at a time. You have to do it in blocks of audio samples at a time, thus a buffer is necessary, and this invariably is going to add latency. Directsound also has to have a buffer as well. You can't add effects, e.g. reverb, echo, etc to just one audio sample at a time, you have to do it with blocks of samples. You also can't send just one sample at a time into the spdif packet stream, you need to send a block of samples.
A driverless software solution has to do all of these in individual steps.
A driver on the other hand can for example process the dsound effects on a given block of samples, then convert that block to AC3, then embed that into an SPDIF packet, all in the same buffer cycle, and then start processing the next one.
The delays caused by this have nothing to do with processing time, but rather is a necessity for how the technology works. It just can't be done instantaneously unless the software can magically travel to the future and know exactly what audio samples are going to be produced before they are actually produced.
A driver on the other hand should be able to do all of that with just one audio buffer.
A little more detail: You can't convert audio from PCM to AC3 with just one sample at a time. You have to do it in blocks of audio samples at a time, thus a buffer is necessary, and this invariably is going to add latency. Directsound also has to have a buffer as well. You can't add effects, e.g. reverb, echo, etc to just one audio sample at a time, you have to do it with blocks of samples. You also can't send just one sample at a time into the spdif packet stream, you need to send a block of samples.
A driverless software solution has to do all of these in individual steps.
A driver on the other hand can for example process the dsound effects on a given block of samples, then convert that block to AC3, then embed that into an SPDIF packet, all in the same buffer cycle, and then start processing the next one.
The delays caused by this have nothing to do with processing time, but rather is a necessity for how the technology works. It just can't be done instantaneously unless the software can magically travel to the future and know exactly what audio samples are going to be produced before they are actually produced.
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
You guys forgot about ASIO and the software works based on ASIO. You don't have to trust me but like I said, if you guys can't even find the software then I doubt that you guys could make it to work because you need to do some one time setting up before you can use it. You need to setup the buffer size, What U Hear volume, ASIO sample rate and some other thing correctly first before you can get it working. You can even find a link the software on Creative forum because the software has been mentioned quite a few times there. What I meant by no latency and no performance loss is the latency and performance loss are so low that they are not noticable, not on my previous X2 3800 and 7950GT setup at least.
@Phide, even a 8800GTX will be the bottleneck at 1920x1200. At a lower resolution however the card could push out a very high FPS that you won't even notice the difference when using a software encoder or a hardware encoder.
Btw are you guys sure that Prelude will use a hardware encoder? Even the X-Fi uses software for DD/DTS decoding. The decoder has been proved to be a software decoder within the driver on Creative forum. Heck, even my OEM Dell X-Fi that didn't have the said "hardware decoder" can have the DD/DTS decoding by using a modded driver for retail cards.
@Phide, even a 8800GTX will be the bottleneck at 1920x1200. At a lower resolution however the card could push out a very high FPS that you won't even notice the difference when using a software encoder or a hardware encoder.
Btw are you guys sure that Prelude will use a hardware encoder? Even the X-Fi uses software for DD/DTS decoding. The decoder has been proved to be a software decoder within the driver on Creative forum. Heck, even my OEM Dell X-Fi that didn't have the said "hardware decoder" can have the DD/DTS decoding by using a modded driver for retail cards.
That's wrong. It uses software based DD/DTS for Vista only due to DRM restrictions. All other Windows O/S have hardware based decoding.
Oh because that is so hard to set up...
...All we are asking for is some proof. So far your proof is "well you guys are too dumb to set it up". Very convincing.alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
Believe what you want as I gain nothing even if I proved it to you. I just want to say that I know it is a software solution, I've tried it and there is nothing wrong with a software solution, not on my X2 3800 + 7950GT setup at least. I also know a few other guys that have tried the software solution and live happily with it.That's wrong. It uses software based DD/DTS for Vista only due to DRM restrictions. All other Windows O/S have hardware based decoding.
Oh because that is so hard to set up...
There seem to be a lot of arguments here about whether the X-Fi provides a noticeable performance boost in gaming or multimedia, but this particular card (the Auzentech) is desirable because it has high quality DACs (and I guess because of the real time DD encoding). If you're one of those people that actually knows what individual DAC chips sound like or you like swapping op amps on modded cards, DACs or head amps, then you may see something you like in this card.
People running decent head-fi or Music or Home Theater rigs off their PCs, people considering this as an alternative to external DACs and the like will probably be attracted to this card. If you just want a theoretical minor FPS boost you could just buy a Creative X-Fi card.
People running decent head-fi or Music or Home Theater rigs off their PCs, people considering this as an alternative to external DACs and the like will probably be attracted to this card. If you just want a theoretical minor FPS boost you could just buy a Creative X-Fi card.
And I just want to say you are full of shit.
What's so hard with providing a link? I'm guessing it's one of two reasons. First and perhaps the most obvious reason, the encoder simply doesn't exist and you used the mystical ac3 encoder to discredit the importance of the lack of ac3 encoding currently on the Prelude. I really hope this isn't true.
But what I think is more likely, that this encoder does exist but is not nearly "production quality". It's probably riddled with bugs, missing features, and performance issues. You're too afraid to show us the link because you know the program in question isn't really in a usable state for anything advanced (like games). So you cover up and say "yeah I know bunch of people who use it and it works fine" without actually providing any evidence.
I don't think we are being too unreasonable by asking you to provide evidence of your claims. And saying "you're too dumb to use it" or "a bunch of people use it with no problem" isn't evidence. I don't care if you show us the link or not, but know no one is going to take you seriously if you keep making outlandish claims without backing them up. You're acting pretty ridiculous right now.
IMO it's more a proof of concept than a solution and I'd only use it if my receiver didn't offer analog inputs, but just to stop this "I know but I won't tell you!" nonsense, here it is for anyone who wishes to give it a try:
http://www.driverheaven.net/general...ocnexk-released-realtime-5-1-ac3-encoder.html
http://www.driverheaven.net/general...ocnexk-released-realtime-5-1-ac3-encoder.html
LOL, it is the kx ac3 encoder, I didn't realise he made it work with the retail drivers now. Yeah I've used this before, of course this has a performance hit and added latency (the developer himself says at times it's up to 200ms!). Oh alg I always knew you were good for a laugh.
EDIT: ah HSE beat me to it.
EDIT: ah HSE beat me to it.
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
I also had a good laugh, I never doubt that you have used it before
:If you haven't used it before, I'm sure that you won't know what is it even after I've mentioned about ASIO and What U Hear
No wonder you know that the program has a high latency up to 200ms and takes a performance hit, you've used the program before [/SARCASM]This certainly isn't always so. Plenty of titles out on shelves are utilizing an extensive amount of CPU horsepower, and there is very rarely a "100% pure bottleneck" in a system, even when you're pushing resolution and graphics settings to the max. The limiting factors can, and do, flip-flop, even with a fairly balanced rig, and modern GPUs are still reliant on the CPU to some degree (this is has been especially so since the advent of the Voodoo 2).
We're not debating about "noticing the difference", either; we're talking about what is actually there. And, again, you're saying there's no performance or latency hit with multichannel encoding without actually furnishing any information to suggest that this is true.
MetalDwarf
[H]ard|Gawd
- Joined
- Jan 20, 2002
- Messages
- 1,072
Can we get back to the Prelude now?
Thank You!
anyways, pulled from the AuzenTech website. looks like June 30th they realized they could not get the card out that day. so now we have another vague "JULY 2007" date. does anyone know if Auzen is partial for paperlauches? or will the card hit retail the day it launches? All in all the delay will probably be worth it to have the superior product and it is also forcing me to wait and save my pennies, July 22 is getting awfully close now. i smell a big upgrade coming.
Current Estimate
* Projected Release date: July 2007
Date of estimate: June 30, 2007
More info: Revised for additional testing of driver and hardware.
Previous Estimates
* Projected Release date: End of June, 2007
Date of estimate: June 4, 2007
More info: Revised due to parts availability
* Projected Release date: End of May, 2007
Date of estimate: April 16, 2007
More info: Initial estimate
Well, surpass anyway. You won't be getting any remotes or breakout boxes, but then again I consider that something of a good thing, as I don't recall having once used either.
$200 or thereabouts would be reasonable. We're essentially talking about an X-Meridian plus an EMU20K1, so it isn't unrealistic.
$200 or thereabouts would be reasonable. We're essentially talking about an X-Meridian plus an EMU20K1, so it isn't unrealistic.
If it works based on ASIO then you can kiss all of your DSP and EAX functions goodbye, as well as making it completely worthless for games, which is thew hole point of DDL, making your software based program completely useless. Why? Because ASIO mandates skipping dsound, openal, etc.
The only games that such a software could be used for would be games that have their own internal sound rendering engine which just sends discrete audio to each sound channel. But AFAIK, no games currently do this. The only game engine that is going to do this that I am aware of is UE3. The only UE3 game out is roboblitz, which aside from being an all around crappy game, only uses dsound afaik.
It's probably best just to ignore him on second thought. A clown won't perform without an audience.
alg7_munif
Supreme [H]ardness
- Joined
- Oct 9, 2006
- Messages
- 5,862
HSE has posted a link to the program but you are still making stupid assumptions without even trying the program out
The program uses ASIO but your game will still use DS or OpenAL. I wonder why some people would have so much negativity without even trying the program. The encoder that only works with kX driver and the newer encoder work differently. Btw I also believe the developer of this program isn't making shit up because I have a personal experience working with him, what I can say is he won't release his work unless he is satisfied with the outcome. Unless you have tried the program and you know what you are doing, all your comments about the program are just negative assumptions and you are more full of you-know-what than I am.If the program outputs true ASIO latency, then it can't output anything that comes from dsound or openal. Those layers come much later in the output stage than any audio that can be considered ASIO compliant. For its purposes, ASIO calls for 2ms maximum latency. In most cases, dsound is going to need much more than that.
The program he linked does use the ASIO mechanism, but it does NOT provide the low latencies you claim. Not even close in fact.
cpu_expert
n00b
- Joined
- May 20, 2007
- Messages
- 2
I talked to Auzentech yesterday on the phone and they told me that production would begin a week from now. So, you should start seeing it on shelves by the end of the month.
MetalDwarf
[H]ard|Gawd
- Joined
- Jan 20, 2002
- Messages
- 1,072
woohoo, about time now lets see if its true
Well there's one last week in July and I haven't heard any new news on the Prelude's availability. I decided to budget in $200 for the Prelude towards my new computer even though it still lacks some features I want (hopefully Auzentech can deliver on their promises). But if it doesn't come out before the end of August, I guess I'll take the Elite Pro out of this computer and continue using that. We'll see what happens.
I decided to budget in $200 for the Prelude towards my new computer even though it still lacks some features I want (hopefully Auzentech can deliver on their promises). But if it doesn't come out before the end of August, I guess I'll take the Elite Pro out of this computer and continue using that. We'll see what happens.