Thursday, June 14, 2018


 32 Synths - 32 minutes - 1 song


Introduction


This is a brief review I did on some of the popular free software synthesizers. The "contest"was simple: I played the same song, consisted of several instruments, using exclusively each of the synths, under a condition: I would have strictly up to 32 minutes to configure the synthesizers. 8 instruments were needed for the song, 8 instances therefore required for each synthesizer, and 4 minutes maximum per sound. Here's the story.

Time to groove


When it comes to contemporary music making and/or sound engineering, for most people the synthesizer is expected to play a dominant role. Professionals or hobbyists are willing to invest an amount of time/money to pick the right tools which will help them to optimize their time and effort spent, in parallel to enjoying a pleasant experience. Comparing the hardware with the software synths, the latter comes with better pricing and computer-wise convenience, often requiring very few compromises in terms of the sound. Modern software synthesizers come in the form of VST instruments and among these, there are some freeware ones that compete easily the commercial synths, both in effectiveness/capabilities (i.e. complexity of signal routing, provided sound resources, etc.) and in terms of efficiency (i.e. efficient coding that means low-CPU consumption, easy preset management and lots of banks).

Although the free synthesizers come at a plethora of shapes and colors and count perhaps close to a thousand, not all of them can be considered as being really efficient tools for music production, allowing for extended sound manipulation at an efficient rate, not overheating your CPU. Although home production would rather not care on occasional overloads, any performance would require absolutely smooth run by the chosen synth(s). In addition, music production with a synthesizer can be seen as a close interaction between the musician and their synth. Efficient communication is therefore a must, and it is true that some synthesizers manage this better to others, allowing for a steep learning curve, or helping starting by providing presets (patches).

In this respect, and after having some (short) experience with freeware soft synths, I decided to perform a trial. Each synth of a group of pre-selected ones would have to orchestrate a song. Such a task is anything but new, however there were additional requirements imposed: there should be 8 instances of the synthesizer, at some point of the song playing all instances together, and there will be only 32 minutes available to choose or shape the sounds/instruments. For this reason, 32 synthesizers, mostly well known, were chosen, each from a different developer. A special song was written for this task, containing some elements, which to my opinion were giving the possibility to better verify the abilities of the synths.

The chosen ones

In alphabetical order, the contestant synthesizers are (in parentheses, the developer):

  • Brainstormer (Roazhon DSP)
  • Charlatan (Blaukraut Engineering)
  • Cobalt (Leslie Sanford)
  • DDX10 (Dead Duck Software) 
  • Dexed (Digital Suburban)
  • Eclipsis (VSTZone)
  • Feldspar (Contralogic Productions)
  • Free Alpha (Linplug)
  • iLoSynth1 (ILoyd)
  • Krolar SX-2 (Kriminal)
  • Latte (Maxsynths)
  • Lokomotiv (Archetype Instruments)
  • Lynx (Xenobioz)
  • Modular (Sonigen) 
  • Noisemaker (Togu Audio Line)
  • Oatmeal (Fuzzpils)
  • Ob-xd (Disco DSP)
  • Odin (Frederik Siepe)
  • PhasOsc (Fretted Synth)
  • Plastique (delaMancha) 
  • Pluton (B.Serrano)
  • Podolski (U-he)
  • RMXL (Krakli) 
  • Sinnah (NUSofting)
  • Subduer (Majken)
  • Synth1 (Ishiro Toda)
  • Tethys2 (TN Audio)
  • The blooo (Full Bucket Music)
  • Tunefish4 (Brain Control)
  • Xhip (Aciddose)
  • WS2 (GTG)
  • Zodiac-6 (Cairn) 

 

Setting everything up

The DAW is Cubase Elements 9.5 64bit, running on an Intel Core i5-7200 @2.50GHz having 8GB RAM using a Focusrite Scarlett 6x6. The 32bit synthesizers were bridged to 64bits using JBridge. So the hardware is rather to the high end for 2018, and the performance of the software synths should be expected good.

Having only 32 minutes to setup 8 instrument means practically 4 minutes per instrument, which can be potentially demanding time-wise, considering the peculiarities of each synth, the (non-) practicality of patch searching, the non-familiarity with some of them, and the difference of the required sounds. Therefore, a good preparation was a must have. Whenever available, the manual of the synthesizer was studied (I consider the existence of a manual, even in some rather primitive form, as a very significant factor to a soft synth’s success) or re-studied, if I had used the synth before, at a level I was feeling somehow confident on the source routing and the structure of the synth. Patch banks were also downloaded from internet, when available, aiming to save time from tweaking knobs (this is another aspect [e.g. to include patch banks], that I believe synth developers should take into consideration seriously, acknowledging though that in some cases the creator of a synth intentionally lets the users to explore it...).

Most of the chosen synthesizers are among the popular ones in the market of freeware VSTis, some even for a period spanning almost 13 years. Some others are less known/popular, either due to the relative higher complexity of their synthesis engine (e.g. phase distortion, additive synthesis), or due to the fact that they are late comers (people sometimes find non-productive to switch from their long-time workhorses). This review/test may give the possibility to the less-known VSTis to demonstrate their comparative skills, against the mainstream elite. But it could also confirm otherwise the superiority of the long-standing choices of the musicians. Or perhaps the line will be drawn somewhere in between. 

The task

I wrote a short original song, called “Exit”, which doesn't claim much credits per se, since I wrote it just to serve this purpose: it includes several melodic lines that alternate or play together, to assist in CPU judgement. It uses 8 different instruments (plus drums), from which some instruments were "standard" and some more specific or abstract, but nothing too extraordinary to prevent completing the setup within 32 minutes. Writing this song also demanded around half an hour (with the aid of Nora CM (free with CM Magazine) and Mildon Strummer (free)). The genre is a downtempo/electronic ballad. As a guide, I have written a first version of this song, using Steinberg's Halion (commercial) and Prologue (commercial), but I decided to not upload that version, since it could give the impression of "right" and "wrong"implementations from the rest synthesizers, especially since I intentionally decided in some cases to deviate from the "normal" sound, when I was feeling that another sound was offering a better expression or it was revealing the personality of a specific synth.

The required musical instruments

The following instruments were tasked to each synth. This is by no means an exhaustive list for the capabilities of a modern synthesizer, but it is rather a typical, representative instrument list, which aims to reveal the personality and the efficiency of a generic purpose synth, given that only 4 minutes will be available to make each sound (or locate the sound or a similar one, if preset banks exist).
  • Bass. A synth bass is expected, with a slight attack resonance and a rather thin body, similar to General Midi’s Syn Bass 1.
  • e-Piano. This is a soft, classic e-Piano high-tone sound, without vibrato.
  • Guitar. It is expected a steel guitar, but if this is not available, or takes time to sculpt, a jazz guitar or an acoustic guitar should be ok, worse case is a triangle waveform with a HP or BP filter.
  • Pad. The pad should have a slight filter sweep, and it should be warm, with a tone close to vocal, preferably non-glassy, alternatively stringy.
  • Accompaniment. This corresponds to a synth line that accompanies/dialogs the e-Piano. It is expected to be a vibraphone timbre with slower attack (bowel-like), and possibly to include some noise.
  • Lead 1. As Lead 1, a typical 90’s brassy lead is expected, which should be possible for all synthesizers. Being responsive to aftertouch is preferred. Unison is a plus.
  • Lead 2. The Lead 2 is the most free-drawn sound. The requirements are only to have a long release, with a sharp attack or zero attack, pretty low decay and medium sustain (e.g. a snapping sound that fades), and it may include a saw-like or PWM waveform.
  • FX. A very basic, white noise “wind” effect, with a filter modulation, is expected. Don’t think such a trivial effect is a given by all contestants, or that it fits the time frame to accurately adjust every filter envelope modulation.

The environment

Drums: The drum line is not part of the competition. The same drums were used in all cases. Steinberg’s Groove Agent (commercial) was used and it was routed to a distortion effect (free). The drums were not pre-recorded, but they were actually playing together with the synth, adding to the CPU-effort reality.

EQ: Each of the instruments uses a 4-point equalizer from Steinberg’s Cubase presets. The same settings were provided to all synthesizers (but with different settings per instrument, of course).

Effects in the mix: The following effects section is applied to the mix (in this order): Steinberg Compressor, TDR Nova dynamic EQ, Steinberg Roomworks (reverb), Steinberg Maximizer, Steinberg UV22HR. The TDR Nova plugin is CPU-hungry, and it was chosen for two reasons: first, the dynamic equalizing gives more chance for dynamics corrections to weaker synths, and saves time from me, and second, the CPU consumption helps emulate a more real mastering environment where more plugins would have been added in the chain (e.g. saturation, stereo enhancer/panner, etc.)

Built-in effects: some synthesizers have built-in effects, which could create a blurred result given the reverb added in the mix. For this reason, the effects were tried to become mitigated. However, the result depended also on the available time and easiness, since some synths don’t have clear ways for quick navigation and meanwhile the clock was ticking. So you should expect that for some recordings, the combined reverbs and delays appear to send some instruments to space. This may be sad, but it is part of this competition.

Multitimbral: All the synths here are monotimbral, to allow for a fairer comparison in terms of CPU. This leaves out some interesting synthesizers, such as, f.ex., OXE, K700 and Anamark.

 

The rating


The following factors were considered during my evaluation:

Fidelity. As you will read, some synths provided all sound solutions in as few as 15 minutes, while for others, the 32-minute limit did not manage to provide a satisfying (to me) result. Note that a sound did not have to appear as an exact copy of a rompler instrument to be accepted – that’s exactly a role for the romplers. However, I consider that the expressivity of a synthesizer can be demonstrated through its ability to (also) emulate some well-known sounds. Therefore, I believe the choice of those instruments should not pose a great challenge in design/patch selecting, for the given time availability. At least, in theory.

Patches. Availability of enough and adequate patches helps the musician to learn quickly the personality and the capabilities of the synthesizer. It also saves time in the beginning, by providing well-made routings. When plenty of patches are available, it is appreciated if there is additionally some organizing.

Manual/Tutorials. The existence of an even primitive manual, and/or tutorials should never be underestimated. Many tricks and tips can exist there, even for the simplest of the synths. It can also provide a MIDI mapping information, contact information, etc.

Design. Even without a manual, a more experienced musician might still be able to find its way to the functions, if the skin configuration allows it. This doesn't mean that a synthesizer should be simple, but it may have a structured way to present its functionality. This leads to better results earlier. 

Efficiency. Finally, the overall efficiency of a synthesizer reflects a quality of its service to the musician, and care from the side of the developer. Remember, that free synthesizers address people who usually cannot afford commercial and expensive ones, and who also most of the times own older or less powerful computers. Distributing to them a free synthesizer which eats a whole of CPU and memory is not the best contribution. It may even be practically useless, which is a pity for the time invested by the developers.

That said, my choice of these synthesizers was made to reflect mostly known and popular ones (e.g. their value has been acknowledged). My scale rates from 1 to 10, with 1 meaning bad, but I would give an 1 or a 2 to a VSTi that is really CPU hungry, or badly designed, or very limited in terms of capabilities, which is far from the chosen "elite". Note that I wanted to include more general-purpose synths from some good developers, but I finally chose to settle to having only one synth per developer. This shouldn't prevent you though from exploring other similar synths from a developer who appears here to have provided a quality VSTi.

 

Results

Here's the outcome of this experiment. The order of presentation is neither alphabetical nor based on my respective ranking, but based on the year of their release (of the version I used). The reason is that I feel it should be acknowledged an added value carried by synths which are considered still capable today without having been surpassed by newcomers; they were probably a breakthrough in the freeware VSTi world, for their time. On the other hand, newcomer synths may have on one hand possibly accommodated the knowledge of the previous successful synthesizers, but on the other hand, they are also expected to be designed to better fit the modern synthesis needs. As a disclaimer, this review focuses on general-purpose synths, and not on synths purposed for specific genres, such as techno, where a dedicated class of wavetable synths/arpeggiators exists today and is ever evolving.

General remarks. Although I was aiming for the same sound(s), each of the synthesizers provided its own version of the song, confirming the distinct personality they have, and therefore their non-redundance. A little bit similar were the versions from the wavetable synths, but then, still the rest of the features of these synths coloured the sound differently. It is noted that the mentioned CPU consumption percentage below concerns the total CPU, that is all 8 instances of the synthesizer, together with Groove Agent in the drums+distortion effect, and the effects in the mix (see previous section for details). For every synth it is possible to listen to the produced song version.

Subduer (2005) - 32bit. Subduer had a very satisfying CPU consumption, between 30-40%. An efficient synth, with a special sound synthesis engine, this subtractive/FM VSTIi should not get unnoticed. It can provide a wide variety of sounds, and it has its own personality. Not a regular synth, but keeping promises of the “regulars” in terms of efficiency. It comes with a very good patch bank, which aims rather to inspire and demonstrate the special character of Subduer. My experience rating: 8/10.
Exit played by Subduer

Zodiac-6 (2005) - 32bit. Unfortunately the CPU was around 60-70%, many times reaching 100%. Not taken as a big disappointment, but the high CPU consumption of this subtractive synth practically meant that this is not a general purpose, all-round synth, for me. Add to this that the sound was not among the best I heard (perhaps somewhat “colder” and more “neutral” than others), this synth appears to be only an optional addition, when a (larger) variation of sound sources is needed, with a given that the song is not at all demanding (in terms of CPU). No manual was available either. My experience rating: 3/10.
Exit played by Zodiac-6

iLoSynth1 (2006) - 32bit. This synthesizer was proven very capable in both its sound and efficiency. As an aged synth, I wasn't expecting any manual - but a brief description of its capabilities was there. It produced a very nice version of the song. The CPU remained constantly between 25% and 55%. The only issue was an occasional crash in my bridged version when opening and closing the instances in Cubase. So, overall it was a 5/10 as an experience rating.
Exit played by iLoSynth

PhasOsc (2006)  - 32bit. A capable and easy-to-use phase distortion synth, PhasOsc didn't make my life difficult due to its few and irrelevant presets. It consumed an average 25%-35% which is very good considering the age of this synth. To its advantage, a comprehensive manual. Sadly, no noise source was available, but a low tuned, distorted note did the job. My experience rating: since I believe the limited number of presets offered hide the true capabilities of this synth, only a 8/10.
Exit played by PhasOsc

Plastique (2008) - 32bit. Some would disagree with the inclusion of Plastique here. Meant to be used as a lo-fi synth, this instrument is expected to be anything but a general purpose synth. As a matter of fact, I intended to include the FMMF from delaMancha, and I resorted to Plastique only when FMMF kept crashing in my DAW whenever I was tweakening the knobs. But Plastique gave its fight, given it's not made for such a task. With almost triple requirements in terms of CPU than most others, it kept reaching a 100% CPU consumption producing spikes constantly. But the sounds were not hard to create, as long as someone would care to disable all the lo-fi switches, that give this synth its distinct personality, but they were unwanted for this song. Interesting experience otherwise, and for this a 1/10 is worth here.
Exit played by Plastique
 
Oatmeal (2008) - 32bit.  A masterpiece of the synthesizer world, the requirments for this song did not really reveal the capabilities of this hybrid synth: Oatmeal is rather an inspiration tool, than just a sound synthesis instrument. A bit harder to navigate among so many patches, but with such an abundance of various banks this task comes not as a real problem. A Low-CPU candidate (just around 30%), as a top-class synth should be, which is maybe still surprising, given the complexity of some of its built-in effects. My experience rating: 10/10. 
Exit played by Oatmeal

WS2 (2008)  - 32bit. Some synthesizers, like WS2, do not reveal their age from their sound. This one was a very good candidate, lightweight and with rich sound and variety.  Since not any useful presets were available everything had to be done from scratch. Nothing to worry about though, since I produced one of the most beautiful electric guitars of this review, an instrument which was proven a problem for most other contestants. The CPU consumption was just 30%-40% confirming that WS2 despite its age, can be a valuable workhorse for the musician. I will add to the Cons that I wasn't able to find any noise source. My experience rating is a 8/10.
Exit played by WS2 

Krolar SX-2 (2009)- 32bit. This is a tiny little surprise. It is a small synth, with clear and nice sound (emulating the Roland SH-2 but in polyphonic), it has good presets, and it is easy to use. It took me only 15 minutes to configure the instruments that I liked. It remained also low in terms of CPU consumption, having a 30%-45%. My experience rating is a 9/10.
Exit played by Krolar SX-2

RMXL (2009) - 32bit. Although not a general purpose synth, but a physical modeling instrument, I decided to include RMXL due to its capabilities. It was proven above my expectations, with very realistic sound where needed, and a CPU that could compare to others. The consumption was mostly around 55%-70%, only slightly above the rest, and only at one moment it reached the 100% - but I keep that pretty none would plan such use to RMXL (e.g. as a generic purpose synth). The general performance, easiness to use, and its presets compensate RMXL. For its melodic, noiseless version that it provided, my experience rating is a 8/10.
Exit played by RMXL

Pluton (2010) - 32bit. A rich – sounding subtractive synthesizer, easy to program and offering many possibilities. Surprising to me that it achieved so low CPU consumption (30%-40%) for the chosen set of sounds, this synthesizer probably has parts developed in C++, and it surely aims to openly compete with other general-purpose VSTis. A bit disappointing that it comes with so few patches (just around 30), with these even not being a collection demonstrating the capabilities and the sound synthesis reach of Pluton. But I believe that someone who would embrace its environment, would quickly develop their own bank, according to their own needs. Still though, I think a large and “tactical” preset bank would have demonstrated better the skills of Pluton. Also to its negatives, the absence of a manual. My experience rating: very few patches, no manual, but efficient and capable, it's a 6/10.

Latte (2012) - 32bit.  A worthy subtractive synth, very able in terms of CPU (it was between 30% and 40%), and with a warm sound, it would only be better with the existence of more patch banks. I found a little bit less easier to design the sound I needed, but I cannot be sure that this was not me.  My experience rating: 6/10.
Feldspar (2013) - 32bit. This is a very powerful synthesizer, and although not belonging to the general purpose synths, it demonstrated a great performance. Mainly aimed for building pads and ambience, it was equally good to carry out regular tasks. It has a limited number of patches, and it may take a while till someone gets the most of Feldspar. The CPU consumption was between 35% to 60% which is very competitive. I couldn't find any noise source, so here's a 8/10 from me.
Exit played by Feldspar

Brainstormer (2014) - 64bit. One of the most pleasant versions was produced here. This synthesizer has excellent sound quality, it comes with a comprehensive manual, and helps the musician find quickly its way into sound creation.  All instruments except the bass were created from scratch. The CPU was also low, between 35% and 60%. Beautiful sound, a lot of capabilities, good presets, and a 10/10 from me.
Exit played by Brainstormer

Eclipsis (2014) - 32bit. A very powerful synthesizer, aimed for pads, but now given the task be a "regular". It managed the task with success. Easy to create the sounds (it's a wavetable, after all), with a few patches and manuals/tutorials. The CPU worked between 40% to 60%, and it took me 25 minutes to finish the sound design. My experience rating is a 10/10.

Free Alpha (2014) - 64bit.  A synth with an almost unbelievable low CPU consumption (20%-40%) for the quality and the versatility of its sound. Plenty of useful patches, easy to pick and start tweakening the knobs. Definitely a synth not to be missed. My experience rating: 10/10. 

Lokomotiv (2014)  - 64bit. Another single oscillator synthesizer (the others are Sinnah and Podolski), which proves itself on the "regulars" task. Very good sound, easy to work with, reliable sound creation. It could be better if it had a manual, but the description of the synthesizer in the download page was rather enough to get you go (especially when you learn that it features a "mathematically generated high definition oscillator without using wavetable or interpolation (sic)"). The CPU consumption was very low, 25%-35% and it featured good presets. My experience therefore does not fall below 9/10.
Exit played by Lokomotiv

Modular (2014) - 32bit. Being the only modular synthesizer of the test, Modular, had to be good. It was better than any anticipation. With extreme possibilities, vast number of patches and well organized, for all kinds of sounds, Modular is actually "teaching" architectures. The sound is rich, and it took me only 20 minutes to fix the instruments. I couldn't find any noise source within these 32 minutes, but I'm pretty sure it's in some module there. This is a very serious synthesizer. The CPU consumption is also very low, a 25%-35%, proving the care taken by the creators. A 10/10 as my experience rating, for Modular, then.
Exit played by Modular 
  
Tethys2 (2014)  - 32bit. This synthesizer is modelled after PAiA Proteus-1 but it also adds features, as far as I understand. The sound is rich, but the interface not very easy to grasp and there is no manual. Still, I needed only 20 minutes to finish with my sounds. No noise source and high CPU further reduce the profile of this synth. Although it consumed 30%-40% CPU most of the times, it had three 100% peaks, producing spikes during live play. Nevetheless, I liked the personality and the sound of this synthesizer, which, to be fair, was not meant to be a general purpose soft synth. I'll keep in touch, and a 4/10 as an experience.
Exit played by Tethys2

Synth1 (2014) - 64bit. What could possibly someone say for this popular synth?  Low CPU consumption (a mere 20%-30%), incredible versatility, absolutely useful and plenty of presets, and you’re ready to go! A winner synth, no surprising that it’s so popular. My experience rating: 10/10. 
Charlatan (2015) - 64 bit. The CPU consumption was an amazing 10% or less, like this synth is transparent. Charlatan is definitely a winner in terms of CPU consumption. A straightforward synth, coming with an abundance of well-organized patches, aims to help quick and efficient music making. It has a bit thinner sound than the rest of the synths discussed here, but who really cares, since you may run twice the synth for a line you’d need fatter sound, without your DAW even noticing it. My experience rating: 9/10. 
  
Podolski (2015) - 64bit. This is a powefrul synthesizer, able for extreme sounds, which when you realize that it uses only one oscillator, you understand that exist many other ways to produce quality sound other than just stacking oscillators. To get to the desired sound was not straightforward however, no matter the large factory banks. The result was still satisfying to my ears, even though I had to spend the complete 32 minutes. To its Pros, the low CPU consumption, that stayed 20%-40% in my case. My experience rating is a 8/10.
Exit played by Podolski

Odin (2016)  - 64bit. Odin is a synthesizer that comes without presets, because it simply doesn't need any. This wavetable/FM/subtractive synth is so simple to use, that building your sound is actually real fun, no matter how complex the sound. Beautiful sound, as expected. It also managed (although many oscillators per instance were active) to remain at a 50%-60% CPU consumption. My experience rating is 9/10.

Sinnah (2016) - 64bit. Perhaps this is the synthesizer with the weirdest panel among the rest here, but it is still easy to grasp the idea and develop the sounds. It's one of the synths hard to classify, but I would consider it as an additive/vector synth. It left me bit undecided on the sound design result, but I can't say that I didn't appreciate its capabilities. Afterall, I had finished after 15 minutes. It is somewhat more heavy than most of the others featuring a 40%-60% CPU consumption, but with a single peak on 100% (which was repated in further replays). My overall experience is a 3/10.
Exit played by Sinnah 

Cobalt (2017) - 64bit. A beautiful synthesizer, that comes with a very friendly manual, and good presets to start producing music. It took me only 21 minutes to get a satisfying sound, allowing also for the synth to reveal some of its specialties. With a low CPU consumption (30%-45%), Cobalt can definitely be a workhorse synth for many genres. My experience rates to 10/10.
Exit played by Cobalt

Dexed (2017) - 64bit. Dexed, as a simulation of Yamaha DX-7 is a pure FM synthesizer. This is good news, because FM synthesis has now become widely known, but it is also bad news, because Dexed is fully compatible with around 3,000 patches from DX-7, that make the knob tweaking not needed, since someone may have access to the original sounds of the '90s. I needed only 20 minutes to decide on the sounds, and the CPU was very low for such a synth, only 25%-35%. A great experience, so a 10/10 from me.
Exit played by Dexed

Lynx (2017) - 32bit. Due to partly being developed in C++, this capable, phase distortion synth was able to easily pass the test of the CPU (staying between 30%-50%). With a very warm and attractive sound, this wave-draw synth also provided a very interesting song interpretation. Its inability to use a noise source (my attempts made via the wave drawing process resulted into sounds that still included a melodic content) can be easily forgiven, for it is very expressive otherwise. My experience rating: with a minus for the noise source absence, but countered with an extra plus for the well-made extended patch banks, a 10/10.
Noisemaker (2017) - 64bit. This synthesizer provided me with one of the sweetest versions. Very easy to use (it took me 25 minutes to design the sounds), with a manual and adequate presets, Noisemaker would only be slightly heavier than others (the CPU was 35%-65%), succeeding though to always stay below the risky zone of 80% (which was not the case for other synths). My score is a 9/10.
Ob-xd (2017) - 64bit.  This is a synth with a rich and warm sound, and it comes with great patches. It is not a really low-CPU synth, but it can compete the rest under some conditions (the CPU was mostly 40-60%, with a peak once to 100% - this is the same performance level as Sinnah above). The single peak to 100% was at the moment of the start of playing all the instruments, and it was not accidental (persistent through several repetitions). But I believe one has to acknowledge the quality of the sound of Ob-xd, and include it in their options (just don’t overdo it, maybe). My experience rating: due to the 100% CPU peak, but also considering that I’ve got one of the nicest song interpretations, a 8/10.

Tunefish4 (2017)  - 64bit. A very good quality additive synthesizer, not easy to learn, but not the hardest either. All sounds were manipulated by me, but I had to start from some of its presets. An excellently optimized code, consumed a mere 15% of CPU, only to be compared with Charlatan, with Tunefish4 providing however a greater sound ability, to my opinion. My experience rating is a 8/10.
Exit played by Tunefish4

Xhip (2017) - 64bit. Xhip is a very versatile subtractive synth, promising a low-CPU (achieved easily a 30%-40% only) and a wide ability in sound design, which came as a little bit of surprise to me. Easy to use, comes with an adequate patch bank, and is definitely a valuable addition to the regular, general purpose synth groups. My experience rating is a 9/10 even without having found a solution for a noise source during design.
DDX10 (2018) - 64bit. The other pure FM of the group (the first was Dexed), is shown to provide a rich variety of sounds. It lacks the number of presets but the factory ones are very good, and may help in further design. This is an easier synthesizer than it may appear to be. With an extremely low CPU consumption (15%), this is a match only to Charlatan and Tunefish4. My experience rating is a 7/10, because although my time didn't let me to provide the best sound configuration, I will not blame the synth for this. I will only reduce a bit from the perfect due to the absence of a manual.
Exit played by DDX10

The blooo (2018) - 64bit. This synthesizer gets advertised for its low CPU consumption (which is true, with a mere 15% CPU consumption in my case) but what is really valuable is the richness of the sound. In other words, the efficient code doesn't come at the expense of its sound quality. I found it a bit hard to design my guitar, but then it is true that I went only quickly through the manual, which reflects to that I needed a full 32 minutes for the sounds. All in all, the abilities with this synth are wide. My rating is a 9/10.

What about the others ?
There are a few other synthesizers/developers which I would like to have included here, but time restrictions did not let me do it. At least, I believe it is worth to mention them: Crystal (Green Oak), Czynthia (Algomusic), KX77MOD-CSE (X.Kalensky), Rainbow (Big Tick) and SixMonthJune (Elektrostudio).

 

About me


I'm a music hobbyist enjoying writing and instrumenting music in my spare time. My involvement with the music begins since 1985. Among others, I bought my Roland SH-101 in 1986, my Yamaha TG-500 in 1992 and my Yamaha Clavinova and Yamaha SY-35 in 1994. I still have (and use) the first three ones, together with my Access Virus TI2, Clavia Nord Lead A1, Novation Peak, Behringer DeepMind 12, Elektron Digitakt, Roland SH-201, Roland JV-1080, Behringer Neutron, Behringer Model D  and my preferred software (VST) instruments, controlled by Samson Graphite 49, E-Mu X49 and Omnitronic FAD-9 (my SH-101 is controlled using Doepfer Dark Link).



What was your favorite interpretation? Your comments are welcome!

Posted by at 2 comments:
Subscribe to: Posts (Atom)