FM3 VS Quad Cortex

[Romo82]

Extensively tested how? High loads of what?

Due to my own gear experiences, I’m weary of anything that hasn’t been put through a snowy winter in New England, where like I said above, EVERYTHING gets covered in dirt and salt. Every bar floor is covered with it, every stage gets covered in it. Unless you’re bringing a fresh pair of shoes to put on before a gig, the shit gets everywhere. I’ve had to replace switches as a result of that in the past and those were just standard switches without any kind of encoders.

R&D is all good but until it’s out in the hands of people and truly put through the rigors of gigging in less-than-optimal climates/situations, I’m not sure how extensive “extensive testing” is.

Exactly. Testing of "100,000 presses without failure" or the like is not testing of drunken foot slamming into the switch from any angle. Obviously its impossible to test every potential scenario out there but then that's why you put a shroud or kick bar on it. The QC switches stand very proud of the chassis to my eyes and with that comes increased leverage.

It really does seem like its designed more for a desktop than a stage top. Which is perfectly fine, but then market it as such.

 

[vika12]

It depends on what you plan to do with your device. If you gig, like I do, you'll want the FM3. It's built like a tank, it's a small footprint, and it sounds great. The FM3 has way more amp models and effects that the QC (based on what I've seen in the QC manual). The "highlight" of the QC is it's capture feature -- however, if you're like me, you have only one amp and a couple of pedals laying around as a backup to the FM3 (I sold off most of everything in favor of the Fractal). Besides, the capture is simply a snapshot of the amp or drive pedal you are capturing; you have very limited ability to edit or tweak settings on the QC after you've created the capture. Therefore for example, if you want to load different amp settings into the QC you'll have to make numerous separate captures. Finally, Fractal has been around for some time and has a proven track record for excellence. Anyway -- my 2 cents.

 

[IronSean]

I could have sworn it claimed to be the most powerful modeler ever when it was first announced. I could be wrong, but I remember the forums lighting up over the marketing. I’m actually discussing this over on TGP right now, where someone said most of Neural’s hype has come from the Neural fans, not Neural themselves. Meanwhile, one of the last plug-ins they released pictured Nolly hunched over a bunch of amp heads as if he took an electrical engineering class and could just look at them and figure out what were the best parts of each to turn into a plug-in.

I'm not sure if they outright lied at the start then fixed it, or just really downplayed the "floor" part of it to not draw attention to it and lead to this very confusion.

They also keep referring to it as "2GHz Quad-Core SHARC® architecture" despite the fact that the accepted way of discussing this is "500MHz Quad Core" because 2GHz sounds more impressive and even sounds like it trumps the Axe FX III which is a 1 GHz Dual core processor.

It's this kind of skeevy used car salesman tactics that ultimately killed the hype for me.

I think you nailed it. As of this writing, the TGP thread is north of 12,800 posts for a thing that hasn't even shipped.

To be fair, they also have thousands of posts anticipating each Helix firmware release Say what you will about TGP, but you can't beat them on Quantity.

 

[Brian Coonan]

I've watched a few reviews/vids on the QC - a couple of them mention Fractal (naturally) and one guy after his own review said that while the QC is good, it's not nearly as good as the Axe III - so - since the FM3 and the Axe III have so much in common tone wise...if I was looking for a floor unit, it'd be the FM3 with a FC6 all the way.

One thing I was curious about - the QC has those rotary encoders which seem cool if you're sitting at a desk tweaking but what about if you're moving around and do that long leg dive to switch channels or something and hit the rotary encoder enough to make it spin while you're playing live? Seems like a potential for something undesirable to happen. I haven't seen the unit in person so not sure if that's even an issue but was something I thought about happening to me, and not wanting it to.

 

[vika12]

... Another issue that concerns me about the QC is it's power cord / supply. It's wall wart with a "barrel" type plug into the unit. I've had similar "barrel" plugs on pedals come loose (even slightly) at gigs and cut power during a song. That's a pants down feeling. As someone said above, the QC seems to be more appropriate for desktop / studio use.

 

[Habuman]

Have no interest in the Quad Cortex. Absolutely satisfied with the FM3. Hell, the AX8 still kicks ass. The days of chasing the latest and "greatest" product that promises to be the end-all for authentic amp sounds is over.

The gear I have works. "I don't need anything - except this, this ashtray, and that's the only thing I need, is this."

 

[Harm]

I'm not sure if they outright lied at the start then fixed it, or just really downplayed the "floor" part of it to not draw attention to it and lead to this very confusion.

They also keep referring to it as "2GHz Quad-Core SHARC® architecture" despite the fact that the accepted way of discussing this is "500MHz Quad Core" because 2GHz sounds more impressive and even sounds like it trumps the Axe FX III which is a 1 GHz Dual core processor.

It's this kind of skeevy used car salesman tactics that ultimately killed the hype for me.



To be fair, they also have thousands of posts anticipating each Helix firmware release Say what you will about TGP, but you can't beat them on Quantity.

Hi,I am a noob at this processor stuff but I would like to understand. You say a 2ghz quadcore is called 500 mhz quad core? Could you explain that?

Thank you!

 

[RevDrucifer]

You are probably not the target consumer, just saying....

I’m a musician who owns a modeler and is interested in the development of them, why would I not be their target consumer? Because I wait until the beefeaters have their fill first?

 

[IronSean]

Hi,I am a noob at this processor stuff but I would like to understand. You say a 2ghz quadcore is called 500 mhz quad core? Could you explain that?

Thank you!

Because the processor they are using is a chip with 4 cores, and each of those cores runs at at clock speed of 500 MHz. So the clock speed is 500 MHz, and there are 4 cores each running at that 500 MHz. Conventionally, that is called a 500 MHz Quad Core processor.

Your intel 10900K is a Deca core 3.7 GHz processor, because it has 10 cores and it runs them each at a base speed of 3.7 GHz (ignoring the turbo boost stuff for simplicity). Your RTX 3080 graphics card is an 8704 core, 1.71 GHz chip, since it runs all 8704 cores at 1.71 GHz. etc.

Neural DSP is "cheating" by adding those 4 500MHz cores together to get "2 GHz of power!" even though the chip still only runs at 500 MHz. Fractal advertises the normal way, showing a 1 GHz dual core processor, which is 2 cores each running at 1 GHz.

 

[Harm]

Because the processor they are using is a chip with 4 cores, and each of those cores runs at at clock speed of 500 MHz. So the clock speed is 500 MHz, and there are 4 cores each running at that 500 MHz. Conventionally, that is called a 500 MHz Quad Core processor.

Your intel 10900K is a Deca core 3.7 GHz processor, because it has 10 cores and it runs them each at a base speed of 3.7 GHz (ignoring the turbo boost stuff for simplicity). Your RTX 3080 graphics card is an 8704 core, 1.71 GHz chip, since it runs all 8704 cores at 1.71 GHz. etc.

Neural DSP is "cheating" by adding those 4 500MHz cores together to get "2 GHz of power!" even though the chip still only runs at 500 MHz. Fractal advertises the normal way, showing a 1 GHz dual core processor, which is 2 cores each running at 1 GHz.

Thank you! I actually learned something today haha. I understand.

Kind regards, Harm

 

[Becoming]

Because the processor they are using is a chip with 4 cores, and each of those cores runs at at clock speed of 500 MHz. So the clock speed is 500 MHz, and there are 4 cores each running at that 500 MHz. Conventionally, that is called a 500 MHz Quad Core processor.

Your intel 10900K is a Deca core 3.7 GHz processor, because it has 10 cores and it runs them each at a base speed of 3.7 GHz (ignoring the turbo boost stuff for simplicity). Your RTX 3080 graphics card is an 8704 core, 1.71 GHz chip, since it runs all 8704 cores at 1.71 GHz. etc.

Neural DSP is "cheating" by adding those 4 500MHz cores together to get "2 GHz of power!" even though the chip still only runs at 500 MHz. Fractal advertises the normal way, showing a 1 GHz dual core processor, which is 2 cores each running at 1 GHz.

Ah, then I assume that each row/input is assigned to a core since they allow 4 separate ones. That is pretty scuzzy way to label it if that's true and goes against all the typical standards to sound sexier.

 

[Mr.Modeler]

Extensively tested how? High loads of what?

Due to my own gear experiences, I’m weary of anything that hasn’t been put through a snowy winter in New England, where like I said above, EVERYTHING gets covered in dirt and salt. Every bar floor is covered with it, every stage gets covered in it. Unless you’re bringing a fresh pair of shoes to put on before a gig, the shit gets everywhere. I’ve had to replace switches as a result of that in the past and those were just standard switches without any kind of encoders.

R&D is all good but until it’s out in the hands of people and truly put through the rigors of gigging in less-than-optimal climates/situations, I’m not sure how extensive “extensive testing” is.

There's a few vids and a few statements from NDSP around this. They used robots to simulate many years of use. They're not new to the pedal world. As for road salt on your shoes, maybe give 'em a quick wipe with a cloth or something? This is no different than any other floor-based pedal including the FM3 or AXE FX III with an FC unit.

 

[Mr.Modeler]

Fair enough but I say you’re under estimating/acknowledging the realities of road use on equipment like this.

I live in a snow belt where salt and sand are the realities of walking outside. I've never had a switch fail on any device. By the time I usually setup and get everything ready, the snow on my shoes has melted and they're pretty much dry. What did you do before the QC? Rack gear only? No pedals?

 

[FractalAudio]

Because the processor they are using is a chip with 4 cores, and each of those cores runs at at clock speed of 500 MHz. So the clock speed is 500 MHz, and there are 4 cores each running at that 500 MHz. Conventionally, that is called a 500 MHz Quad Core processor.

Your intel 10900K is a Deca core 3.7 GHz processor, because it has 10 cores and it runs them each at a base speed of 3.7 GHz (ignoring the turbo boost stuff for simplicity). Your RTX 3080 graphics card is an 8704 core, 1.71 GHz chip, since it runs all 8704 cores at 1.71 GHz. etc.

Neural DSP is "cheating" by adding those 4 500MHz cores together to get "2 GHz of power!" even though the chip still only runs at 500 MHz. Fractal advertises the normal way, showing a 1 GHz dual core processor, which is 2 cores each running at 1 GHz.

And that's ignoring IPC. The Keystone DSP used in the Axe-Fx III does at least twice the IPC (probably closer to 3-4x).

Here's the BDTI benchmarks:
https://www.bdti.com/MyBDTI/bdtimark/chip_float_scores.pdf

The Axe-Fx III uses a dual-core TMS320C6657. The SHARC+ used in the FM3 and the QC are similar in performance to the ADSP-213xx. Same basic core. Slightly higher clock speed and some accelerators. Draw your own conclusions. The SHARC+ is not a bad chip but it's nowhere near Keystone levels of performance. It's not even as good as the old TigerSHARCs which they discontinued. It's designed for cost-sensitive consumer applications.

 

[Douglas Costa]

And that's ignoring IPC. The Keystone DSP used in the Axe-Fx III does at least twice the IPC (probably closer to 3-4x).

Here's the BDTI benchmarks:
https://www.bdti.com/MyBDTI/bdtimark/chip_float_scores.pdf

The Axe-Fx III uses a dual-core TMS320C6657. The SHARC+ used in the FM3 and the QC are similar in performance to the ADSP-213xx. Same basic core. Slightly higher clock speed and some accelerators. Draw your own conclusions. The SHARC+ is not a bad chip but it's nowhere near Keystone levels of performance. It's not even as good as the old TigerSHARCs which they discontinued. It's designed for cost-sensitive consumer applications.

In fact you worked some miracle being able to run most of Axe-FX III code on FM3 hardware.

 

[LosD]

And that's ignoring IPC. The Keystone DSP used in the Axe-Fx III does at least twice the IPC (probably closer to 3-4x).

Here's the BDTI benchmarks:


The Axe-Fx III uses a dual-core TMS320C6657. The SHARC+ used in the FM3 and the QC are similar in performance to the ADSP-213xx. Same basic core. Slightly higher clock speed and some accelerators. Draw your own conclusions. The SHARC+ is not a bad chip but it's nowhere near Keystone levels of performance. It's not even as good as the old TigerSHARCs which they discontinued. It's designed for cost-sensitive consumer applications.

Interesting that they went with SHARC+ instead of Keystone. Maybe cost related or perhaps around code development complexity on the Keystone end?

 

[Romo82]

I live in a snow belt where salt and sand are the realities of walking outside. I've never had a switch fail on any device. By the time I usually setup and get everything ready, the snow on my shoes has melted and they're pretty much dry. What did you do before the QC? Rack gear only? No pedals?

I've run both rack and floor setups but you're focusing on a single point of my overall issue with the design of the QC in having no guards against foreign objects striking the unit. For all we know the switches could be the finest piece of engineering the switch world has ever seen (which I doubt) but the very fact that there is no attempt made to offer even the slightest bit of protection to the unit is my concern.

 

[Nick Barba]

In fact you worked some miracle being able to run most of Axe-FX III code on FM3 hardware.

Is there any future thought to a sort of compatibility mode where the FM3 will run an Axe-FXIII patch but in a scaled down manner to fit the processing capability of the weaker unit? Could see it being a huge time saver and there might not need to bring full AxeFXIII to a show if you can just load it up quickly on FM3 and sound is 95% there.

I ask cause I have a QC on preorder but this feature may make the FM3 a better choice for me

 

[vetteboy79]

Agreed on all points.

My buddy is in the market for a modeler and LOVES my AxeFX III, but he certainly doesn’t need all the bells and whistles. I’ll be dialing in 95% of his presets anyway. I told him to just get an FM3 and be done with it, but knowing how he’s not exactly technologically proficient, I thought the QC might work out better for him since it’s drag and drop. However, until those things have been in the hands of user’s and tested for durability, I won’t advise anyone to drop $1600 on them.

My first questioning about it was based around those encoders. When gigging in New England during the winter, you’re not going to be able to prevent dirt/salt from getting all over it unless you keep it behind you and use a MIDI pedal to control it. I had switches fail on my GCPro and I’m assuming it was due to the dirt/salt getting in there. Never mind beer and booze getting spilled on it. And it’s funny, whenever that gets brought up, someone inevitably asks, “How often does that happen? I’ve never had that happen once!”.....well, ya need to play some more VFW’s and dive bars. If someone from the audience wasn’t spilling something on my board, I was.

I could imagine one good rum and coke on those encoders would do a great job at gumming them up once it dried.

I've had my AX8 since 2017 and it's been played all over the country, from street gigs, VFWs, dive bars, to festival stages. It's pretty gross, to be honest. But it's been absolutely bulletproof, and for sure the most robust of any other floor modeler brands. Its form factor works great for what we do and I have no intention of retiring it anytime soon, hell I don't think I've even editing anything on it for a few years now.

I have the FM3 on my studio desk at home for doing session work, fewer buttons is actually a plus for me in that sense...I don't use the 'foot' switches at all. I wouldn't hesitate to gig with it but I got it specifically because I was tired of unpacking my AX8 road rig any time I wanted to record at home. Yes the FM3 sounds better, but I feel that capability is better used for session stuff where the nuances are more noticeable, and keep the bombproof "almost as good" rig on the road where no one will notice anyway.

I have the Nolly DSP pack which I was using in the interim before getting the FM3, and I haven't touched it since. I think I did one session project with it and had to do a lot after the fact to get it sitting well in the mix. With the FM3 I rarely go beyond the factory presets (aside from turning off most of the reverb/delay stuff) and they just work great out of the box.

 

[RevDrucifer]

I live in a snow belt where salt and sand are the realities of walking outside. I've never had a switch fail on any device. By the time I usually setup and get everything ready, the snow on my shoes has melted and they're pretty much dry. What did you do before the QC? Rack gear only? No pedals?

If they were regular switches, I wouldn’t bat an eye, but the fact that they’re encoders that have to have some degree of accuracy, well, things like that aren’t normally stomped on. It’s an AWESOME idea and I hope it goes off without a hitch because I’d love for that to be utilized with other brands/products. I’m just looking at it from the POV of someone who has HAD switches fail.

Almost every gig I played in Massachusetts had you loading in right before you got onstage, so straight from the door to the stage. If you were lucky, there wasn’t a narrow staircase built in 1890 you had to walk down with a 4x12 to get to it.