george

Yes, you can control the e28 hardware volume with Roon Remote on Mac in ASIO mode. Roon will enable volume control from Roon Remote to Roon Core to RAAT endpoint. From that point it is a question of Roon/HQPlayer integration. You need to check with them.

Thank you, Ted. The way we do DSD is ASIO for Windows. We will follow your advice. No need to mimic - we have our own stuff. Best, George

barrows, 4est, Yes, we are working on OS X drivers. We had to shift attention for a while to work on another exciting development - DSD support. This is where the action is - We enabled DSD for existing customers with a software update. Regarding the price of exaU2I - we do all our driver, hardware and firmware development in-house in order to achieve unrestricted performance and quality. We aim to be the best and we will not compromise quality for price. Most of our users use only two channels. They appreciate the sound realism, detail, dynamic and the overall stability of the device. Users posted in various forums that exaU2I delivers excellent value even for two channel use. To answer your other questions - the clocks are located side by side on the board - the one that you don't recognize is black. We don't use a processor and therefore we don't need a third clock. The FPGA technology gives us fast parallel execution and much less jitter. Regards, George

barrows, Jitter should be measured on the receiving end. Not on the SPDIF or I2S device terminals, but after it is received on the DAC end. The problem here is that there is no dedicated clock line with SPDIF - timing mist be recovered from the data. It is possible to recover to some degree the clock with costly solutions. However why loose it to have to recover it again? Let say the clock is recovered, you still have the bandwidth limitations of a legacy interface. I am making a point, not claiming that all devices are born equal. My point is that ES9018 properly setup for SPDIF is an inferior solution to ES9018 properly setup for I2S. BII users can decide for themselves. When the firmware is removed from a BuffaloII DAC, it works with the chip default setting and it becomes rock stable with I2S source. (volume control is lost for this test - be careful) Regarding the oscillators, the information on our website is correct. What we don't have on the board is processor.

Sorry to intervene in a user discussion, it is misleading and incorrect to say that a compromise is necessary to use exaU2I with a Buffalo II DAC. Buffalo II default firmware settings are optimised for SPDIF use. Several third-party I2S devices have been found incompatible with these settings. The dropouts are not caused by jitter or "noise on the I2S lines". Simply I2S uses frequencies 64 times higher than SPDIF. Here is a reality check - SPDIF has jitter levels measured in nanoseconds. I2S jitter is usually picoseconds. How come hundreds to thousand times lower jitter is causing instability? Using I2S sources with Buffalo II requires settings optimised for I2S. The result is not inferior in any respect because of the benefits of much lower jitter in the source. There are numerous benefits: higher sampling rate, 32bit precision, bit-perfect asynchronous operation, bandwidth sufficient for multichannel operation at 384 kHz, the ability to play DSD64/128. This discussion is not relevant anymore. Buffalo II is discontinued and Buffalo III is out. Buffalo III offers excellent flexibility so users can experiment and get the best performance without replacing the firmware. Regards, George

Peter, Hopefully we are now done with the emotions and the accusations. Your energy will be better spent to make your technology more valuable instead of throwing mud at the alternatives. I would like to come back to the essence of this discussion. I started with a proposed solution in principal for ultra-high-end multi-channel. My extreme case has evolved to more flexible consensus: Implementing stereo or multi-channel by starting with a bit-perfect baseline configuration and applying processing if needed. The advantage of this approach is that it makes is easy to evaluate the benefits and trade-offs of using any enhancements. Any particular scenario is different and there is room for any taste. I would encourage people with access to good listening rooms, full-rage speakers and analogue amplifiers to try setup with minimized processing. Please come back with your own test results and experiences. Besides the principals that I've expressed, another hot potato was my DIY toy. It offers a clean, transparent high-performance data transfer, so it is equally useful for people that need more processing. exaU2I will not degrade or tint the output of sophisticated upsampling, digital crossovers, room correction or XXHighEnding. These topics deserve another thread. I will be glad to offer support and knowledge about the exaU2I interface if asked. Cheers, George

Hi Peter, You are right about the dream. It is a dream-come-true. I don't understand why you are so bitter about this bit-perfect low-jitter ultra-high-quality stuff. XXHighEnd has a good following. It is a brand based on secret formula, it sounds great, and it makes many people happy. It is Coca-Cola. I am not saying that three is no excitement in sweetened products. What I am preaching here is pure water. Unlike Coca-Cola it has the ability to satisfy thirst. It doesn't make you ever thirsty. Maybe that's why you are so afraid of it. I am happy with my system. Others have stated that they achieved satisfaction with their bit-perfect surround. You are upset. There must be something here. By the way, I have DIY customers from Norway that use XXHighEnd with ultra-high-end horns, and also use exaU2I with Foobar at 384kHz/32bit. I will invite them to this discussion. Or maybe somewhere else. This thread here is about principles, not about products. I have no interest to derail this discussion into the field of subjective preferences and bias. -George P.S. What Google Group? I haven't done anything with Google Groups ever.

Yes, this is possible. I've tested active crossover configurations with Foobar and Allocator. I don't have proper speakers for this test, so my experiments were limited to DAC/software configuration. You can see some screenshots here: http://www.exadevices.com/Blog/tabid/253/articleType/ArticleView/articleId/20/categoryId/3/Using-Allocator-with-exaU2I-and-Foobar-2000.aspx

I think you will have all this in 2 years:)

Hi Wap, My board is usable for a limited audience of people with experience in electronics and audiophile dedication. It is not like "don't try this at home" but it is a different game. You need a bit-perfect ASIO compatible player like J.River media center, Foobar2000, AlbumPlayer etc. I provide a high-performance ASIO driver that is independent from the Windows sound system. So it is like streaming over USB. Sound stream is send to an external device - the exaU2I board. The exaU2I maintains a buffer. When Windows or the USB interface experience an interruption or delay, the output stream to the DAC remains smooth. Finally the output to the DAC is re-clocked with one of the two precision crystals, so it is almost jitter-free. To make it 8 channels you need to attach four Biffalo II or AckoDACs. These devices are not "user friendly", when you buy all the power supplies, output stages, volume controls etc. you end up handling some complexity. The reward is 32bit / 384kHz studio-grade playback. I have to figure out how to make all this simple. I want to say it again - my implementation is a proof of concept. This thread was started to discuss the principals. They work with a streaming devise like the new Oppo or with any true asynchronous USB or FireWire DAC. Several users here have shared their experience with bit-prefect multi-channel implemented in various ways. When you setup your listening room like a studio it sounds like a studio. -George

I am sorry to see you disappointed, Paul. I understand that DIY is not an option for you and I don't advocate it here. However I stand behind the big promise of bit-perfect low-jitter multi-channel. Thank you for bringing your point of view to this discussion. -George

Bleedink, Your post is right in the center of the discussion here. This is exactly the point - to play multi-channel without compromising stereo; Eventually to play multi-channel better than stereo. I own a NAD receiver, an Oppo player and an emu1820m interface. I can repeat the same results with all of these technologies when I apply the principles stated in this discussion. -George

Paul, I think it is the opposite. I've been very careful to keep the discussion clean from hardware and software bias. Throughout this thread I used references to specific products only when I was accused of dreaming. My solution is not commercially applicable to the audience here but it is a proof of concept. My solution does not exclude fancy processing either. Again it is the opposite. It is an opportunity for people interested in up-sampling and high-level DSP. -George

The solution for computer based multi-channel high-end is hard to find. It is also hard to understand what pollution comes with unnecessary processing from the operating system layers and the sound drivers. Consumer gear is overloaded with features that bring processing. Professional devices can be hard to setup and use because they are designed and optimized for a different purpose. I couldn't find an accessible solution for bit-perfect low-jitter multi-channel ultra-high-end PC playback, so I decided to build the missing pieces of the puzzle. The first missing link for me was the computer-DAC interface. It is not enough for the interface to be bit-perfect. It has to be also low-jitter, or jitter free. A PC system brings significant jitter pollution. Sources of jitter are the computer power supply, the lack of precision in the interrupt system, the non real-time nature of the operating system. Additional sources of jitter are interfaces like SP/DIF and unfortunately HDMI. Both interfaces lack clock lines suitable for jitter free reproduction of Audio. A decent computer-DAC interface must resolve the following challenges: • Bit-perfect operation • Jitter-free operation - eliminate computer jitter and jitter introduced by the interface itself • Isolate the computer ground from the audio system ground to avoid hum and common noise. Computers often don't have isolated ground like audio gear. When you put together cable TV, computer and analogue amplifiers, the ground loops become a real issue • Provide enough bandwidth for up to, let say 8 channels. Some people will ask for 16 channels to do active crossovers. Some are happy with four channels. • Be able to play the rare but increasingly popular master files available on the Internet. For this you need the capability and the bandwidth for 32bit / 384kHz The solution for all these challenges is a custom software-hardware solution designed from the ground-up for transparent and dynamic sound reproduction and nothing else. I couldn't find it so I've built it. My solution is not fancy by it solves the issues that I've outlined. It is a DIY thing. The real benefits from bit-perfect jitter-fee really explode in front of you when the last constraining factors are removed - driver and computer-DAC interface. -George

Hello ted_b, It is a big relief to me that I am not the only one experimenting with true lossless multi-channel Hearing is believing; I am surprised that the idea is rejected without experimenting. -George