Feasibility experiment upgrading OASYS to Intel Core 2 Duo

[sebbytriton]

Remember that performance are limited by software...

I don't think Korg every confirmed or denied this. It's just speculation at this point.

Mark

One day, Dan tells that about 20% or 40% of processing can be used by effects. The total of CPU charge is precalculated with a % ratio for each of the effects. It is independant of the processor, this means that with a core 2 duo, the same limitation is applicated because there is no pre-calculated processor usage.

[jerrythek]

I do believe we have stated many times that you couldn't just change out the processor and expect any performance benefits unless the whole OASYS software system was optimized for that processor. Not Linux codes and drivers, but the whole synthesis system.

So while this may be a fun diversion for you, it is relatively meaningless.

Regards,

Jerry

[rkarlberg]

Thanks so much for letting us know about this. I too am very concerned about what will happen to my $10,000 investment in the upcoming years. We have no idea what the future holds, not even the vaguest roadmap from Korg.

Certainly the updates so far have been great, but eventually the hardware will need upgrading, whether in 5 years or 10...

[elvisjohndowson]

I do believe we have stated many times that you couldn't just change out the processor and expect any performance benefits unless the whole OASYS software system was optimized for that processor. Not Linux codes and drivers, but the whole synthesis system.

Actually, Jerry, the Core 2 Duo architecture would run atleast 40% faster and 40% cooler compared to a Pentium 4 processors of the same frequency.

The OASYS application and the various synthesis engines are probably separate processes, co-operating together to form the complete system and would benefit from multiple execution units (i.e. CPU cores). You don't need to do any further optimization to the code to immediately gain these benefits. You will just have to enable multicore processor support in the linux kernel to immediately take advantage of the multiple cores.

What you say about optimizing the code for a particular synthesis engine for a processor is also true, but that can come at a later stage. The instruction set for the last few generations of Intel CPUs have still remained at SSE3, so any software code fine tuned and optimized for the SSE,SSE2 and SSE3 instruction set can run unmodified on the current Core 2 Duo processor. The application will simply benefit from the newer processor architecture that executes those instructions.

So, in the short term, if the linux kernel can be updated to support newer intel chipsets and multicore CPUs, you will surely be able to gain at the very least, twice the processing power currently available to you on the current processor.

Elvis Dowson

[elvisjohndowson]

So while this may be a fun diversion for you, it is relatively meaningless.

For sure, , I debated on whether I should try this, and knew from the start that it probably wouldnt work. But, I couldn't help myself, there was a small chance that the chipset might have been legacy compatible with the linux 2.4 kernel, so I decide to go ahead with it. It took a bit of careful planning and research over a day to find the correct combination of parts and compromise with what's available in the market today.

What was alarming was that the combination that I ended up, itself was rare and not easily available in the market as of 2007, and was rapidly getting outdated. For example, in the microATX form factor, the trend of having onboard serial ports (and that also two serial ports) is rapidly declining and we might not see it next year. Also, IDE is being rapidly replaced by SATA drives and for the moment, you will continue to see IDE and SATA side by side, but in two years most drives will probably be SATA or a newer technology. If you upgrade the internal platform to a microATX motherboard that supports the Quad Core 2 Duo platform as an upgrade kit, that would give it enough bandwith for the next 3 to 4 years.

Best regards,

Elvis Dowson

[Francois]

At some point, it would make sense to keep the Oasys line moving on and introduce a faster machine based around a modern mobo/chipset/cpu, if only to create a marketing impetus for a V2 line.

Secondly, create a "back to shop" upgrade path. Somebody who' s spent $8000 on a machine is not going to say no to a $500/700 upgrade to stay up to date and benefit from more voices, larger banks (if 4Go ram mobo), etc.

[MrT-Man]

Thanks so much for letting us know about this. I too am very concerned about what will happen to my $10,000 investment in the upcoming years. We have no idea what the future holds, not even the vaguest roadmap from Korg.

Certainly the updates so far have been great, but eventually the hardware will need upgrading, whether in 5 years or 10...

I don't understand. Why will the hardware need upgrading? The Oasys isn't a PC, it's a synth that happens to be based on PC components. In 10 or 15 years from now it'll still make really nice sounds when you sit & play it (just like my 16-yr old SY99 does).

Your ears aren't going to get any better -- an amazing patch will still sound like an amazing patch. Korg will NOT be releasing an update to the OS called "Oasys Vista" that you will be forced to upgrade to, and that will require you to get a CPU that is 4x more powerful...

[Drew FM]

Thanks so much for letting us know about this. I too am very concerned about what will happen to my $10,000 investment in the upcoming years. We have no idea what the future holds, not even the vaguest roadmap from Korg.

Certainly the updates so far have been great, but eventually the hardware will need upgrading, whether in 5 years or 10...

I don't understand. Why will the hardware need upgrading? The Oasys isn't a PC, it's a synth that happens to be based on PC components. In 10 or 15 years from now it'll still make really nice sounds when you sit & play it (just like my 16-yr old SY99 does).

Your ears aren't going to get any better -- an amazing patch will still sound like an amazing patch. Korg will NOT be releasing an update to the OS called "Oasys Vista" that you will be forced to upgrade to, and that will require you to get a CPU that is 4x more powerful...

I couln't agree with you more. It is my opinion that the Oasys was never intended (thank God), to be involved in the 'my processor is bigger than your processor' game. I understand that there are certain physical restrictions, such as number of banks but, there is a long way to go before I ever fill that 40GB HD. If Korg continues to provide the caliber of updates and customer support I have experienced so far, I will be forever happy with this instrument.


Andy

[Francois]

Mr T and Drew, you've got a point. However, look at the E-Mu E4 line.

The E4 I bought in 1997 was not the same machine as the E4XT Ultra I now have. And yet, it retains the same OS, architecture, filters... but being equiped with a faster CPU enables faster processing.

What's wrong with the Oasys being upgraded along the same line to provide more ?

[danatkorg]

I do believe we have stated many times that you couldn't just change out the processor and expect any performance benefits unless the whole OASYS software system was optimized for that processor. Not Linux codes and drivers, but the whole synthesis system.

Jerry is completely correct. The OASYS system is highly optimized for use on the current hardware, and as we've noted does several levels of resource management to avoid the "CPU overs" common to native DAW systems.

Without changes to the main OASYS software, swapping the processor will not significantly affect the realtime performance of the OASYS synths and effecs.

- Dan

[MarkF786]

For the current incarnation of the Oasys, the existing hardware is more than adequate for me. For future versions of the Oasys it would make since for Korg to use newer technology, but I'm not hopeful that Korg will release an uprade for existing Os; it would be difficult to manage and I don't see the financial incentive for Korg.

On another note, I wonder if when Jerry says that the software was "optimized" for the hardware, it means that there are hard limits set in the software. For example, the maximum polyphony might be hard-set, not necessarily with the intention to limit future upgrades but to provide a more stable system.

[rkarlberg]

I think the point of this experiment was to see whether a hardware upgrade is possible at all. And it seems that it is. Certainly it would have to be an official upgrade from Korg, with new software optimized accordingly. But most other keyboards are not upgradable at all, so this is good to know.

We just don't know whether this makes sense for Korg, or if they will eventually come out with an Oasys 2.

[tritex4]

It's a big assumption that Korg will introduce another CPU based keyboard.
Going foward, I'm sure Korg will weigh their options in this regard.
It may be a little premature to envision CPU based keyboards as the wave of the future.
Has any of the competition given any hints that they'll follow suite?
The OASYS may be a one time instrument, so this talk about processors and motherboards may be a mute point.

[elvisjohndowson]

Has any of the competition given any hints that they'll follow suit.
The OASYS may be a one time instrument.

Korg certainly didn't follow the beaten path !

[rkarlberg]

We do know that Korg invested a huge amount of money and time in the development of the Oasys. The brilliant part of this is that Oasys technology can be used in many other products. Based on that I believe CPU based products will continue to prosper.

Korg does not have to reinvent the wheel, they can create an Oasys 2 based on what has been learned so far and leveraging existing technology. The software is already there, it would just need some updating to take advantage of the new hardware capabilities. But they could also decide that an M4 would be more profitable.