Discover
learn. create. do.

How TSE Audio Created One of the Best Amp Sims on the Market

by Finn McKenty
music & audio
x50 main

As a DIY musician who lives in apartment, I’ve come to realize that amp sims are my best friend, or should I say, my neighbors best friend. Over the years, I’ve tried just about every sim under the sun, but the most inconsistent has always been the 5150 (especially for OS X). All that changed a couple of weeks ago when I bought the latest version of x50 from TSE Audio, which is hands down the best 5150 sim I’ve used. I’m also a big fan of several other products by TSE including the TSE 808 (their Tube Screamer sim) and BOD (Sansamp pedal sim), so I figured I’d hit up  their lead developer and CEO John Johansen to learn more about x50 v2 and what TSE Audio has coming up!

First of all, it seems like everybody is loving x50 v2! What have you been hearing from users?

We’ve had positive feedback from our users. I have an impression that most of them fell in love with the sound and it’s intuitive interface almost instantly. As this is my first commercial product, I was very anxious about the impression people would get of it, but I was relieved to see it become very popular just over night. People seem to be impressed with the accuracy of the simulation and the feel of this amp at first, and then they try it out in a mix and are even more amazed. Of course there will be people that won’t like it as much as others, mainly because the sound doesn’t fit them. Some people just won’t like it because it is no longer freeware and it requires them to purchase it. As a sidenote, people are very pleased with our customer support too 😉

It’s been quite a while since the release of x50 v1, which was just an amp sim. Can you tell us what made you decide to turn it into the “suite” in v2? What are the sonic differences between v1 and v2?

My first intention was to make a simple amp sim, but then I thought “wouldn’t it be cool to have the TSE808 in the same software?” I implemented it shortly after. In the middle of my development I got in contact with Pavel Shevtsov, my current graphic designer, and I was struck by his talent right away. His work gave me more ideas and I ended up adding a IR loader. I had been doing some research on speaker modeling and the option that I could add an extra flavor to it in the form of the “movement” algorithm made me go through with it… and then it was basically a suite 🙂

It wouldn’t be complete without making a rack with effects. I spent quite some time on the tuner which was my first addition to the rack. I’m very pleased with the accuracy you get for low-tuned guitars and even bass-guitar. It reacts like a hardware-tuner and the user feedback on it is great! The Delay and EQ are custom made as well and has it’s natural place in this amp-setup.

The sonic differences between v1 and v2 are huge in my opinion. I re-modeled the entire amp from scratch with new algorithms to interconnect the gainstage models. Example: Pushing the grid of a tube into saturation now loads down the previous stage etc, the capacitors will charge and as a result the working-point of the tubes changes and you won’t get that typical “flat” sound older amp sims are so famous for.

It’s simply has more life in it now, the low end sings better and the high end isn’t as harsh as my first version, it cuts though in a mix better because of this. The new algorithms made the sound warmer also. If you monitor through each of the X50 versions while playing your guitar, one at a time, you will notice the dynamic difference instantly.

x50 speaker

x50 v2 now includes a cab sim component (IR loader). You can choose from several dozen included IRs, or load in your own.

I really love the stripped-down, easy-to-use approach you took with x50 v2. It has everything I want (overdrive, amp, cab, and EQ), without a bunch of useless options and effects I’ll never use. What made you choose this approach?

I wanted to make this software somewhat untraditional, and make it suit most people without confusing them with too many choices.

Some people want it all in a huge package, and when they get it they some times can’t get their head around learning to use it, so having that in mind we ended up with a good sounding and easy to use final product, I think it was a wise decision in the long run.

An intuitive and easy to use GUI was also a big priority when designing this, it also leaves room for future expansions 😉

It seems like there are less 5150 sims compared to other high-gain amps like Mesa, Engl, etc, and many of the 5150 sims that do exist are not perfect. Is there something about the 5150 that makes it especially hard to simulate? Why did you choose to do a 5150 sim?

The 5150 is a huge amp, in terms of modeling at least. With the “analog approach” – where you write down the equations for each electrical network and simulate it in your software, it has some stages that are particularly hard to emulate such as cold biased stage, local feedback etc. the main reason for this is numerical instabilities when doing the real time calculations. These stages lies right in the most delicate part of the distortion-shaping and is important to get right (in my opinion) to get a good sounding model with the right dynamics.

I ended up making a 5150 sim because of the challenge and the complexity of this amp. I had just finished the X30 (Engl e530 sim) and wanted a bigger project, even though it has felt too big from time to time 😉

Some great tips on tweaking x50 v2

Can you talk a little about your overall design philosophy for x50? Do you take the approach of modeling each component (like Bias or Axe FX), or just create a new design that “sounds right” even if it doesn’t model the components (like Pod Farm)? 
I use the “analog modeling” approach, where you simulate the actual circuits. In my opinion, this has the best sound when done right and I think it is most the flexible. The benefit of this approach is that you could add mods and other cool stuff by simply changing the values of components 🙂

In some scenarios you could simply remove parts of the amp and mix in new components to make a completely different amp.
The downside of this approach is that the computational complexity grows with the size of the amp, so the hardest part is actually to optimize it afterwards for speed, to make it usable for the general public 🙂

What’s next for TSE Audio? Can you give us a sneak preview of  your upcoming projects? 

I still have some ‘holes’ to fill in the new X50 project. We have been talking about adding an extra amp and some additional effects as a bonus for our existing customers.

Personally, I want to make the software modular so that it can add and remove effects on the fly, but that involves some rewriting of the software. It’s a big job.
Also, I haven’t made any clean amps yet, so that is one of the things I would like to do for a future project 😉
Like TSE Audio on Facebook to get the latest news.
Tags: , , ,

Related Articles

Comments

Finn McKenty

Finn is the producer of CreativeLive’s audio channel.

You can email him at finn [dot] mckenty [at] creativelive [dot] com

@finn_mckenty