The Agent’s 10-stage Process

The Essence Agent uses a 10-stage process to automatically generate, adapt, and/or optimize computer-readable instructions for parallel processing in response to its evaluation of memory latency, number of data fetches, number of instruction cycles, order of instructions in the instruction pipeline, size of the cache, wattage consumed by each series of instructions, and the modes of the underlying chip and instruction set architecture chosen.

Power00: Bootup & Spawn

Operating System(OS) specific process-launching/thread-spawning or Hardware UEFI/BIOS specific ( or both ) for startup process

Power01: Resource-Assessor

Handles initial processor/bus/memory assessment and resource pooling

Power02: User-Setup

Identity Confirmation and Location Probing ( find current user, their last Essence session, along with style, rules, and history )

Power03: Authenticator

Manages a variety of verifications, guards, and runtime trust-checks. Only handles the other 7 Powers and itself but does so natively at startup.  Generally built to use any Chip-embedded block-encryption, such as the AES or TPM features on many modern CPUs or other chipsets. Handles verifying the contents of the Agent haven’t changed, that several classes of numerical solvers work as expected, and that various digital signatures, such as hash-verifications, are met as expected.

Power04: Instruction-Maker

Matches streams of Essence Elements into foundation segments of hardware-machine instruction or emulation instruction for 1 or more boot-time chips such as CPUs, Bus-controllers, DSPs, Media-Codecs, & other chips available at startup or uniquely configured (SoC). This Power remaps the existing instruction-set, special conditions and other inter-dependencies for runtime instruction generation for the particular hardware & software stack ( such as ‘system calls’ in an OS ). We do not need a compiler to generate machine instructions.

Power05: Executor

Handles connecting a series of ‘Code-Maker’ results with other Powers from other Wantverses based on the platform’s prior runs and training/history data with ranges of scored (good/neutral/bad) results.

Power06: Referee

Manages issuing, assessing, and balancing various results of the Power04:Instruction-Maker running across the currently available chip.

Power07: Terminator

Manages all shutdown or unrecoverable error states, blackbox logging, & cleanup issues specific to the Platform.

Note: Each Power above has ‘hard-coded’ instructions along with a knowledge-database and supporting data that has been trained and tuned for the specific platform, its generic and unique features and all the caveats that must be observed to optimally generate, run and evaluate machine instructions on the fly.

Next is Morpheus and Maestro

Introducing Morpheus^TM

Given we expect that some will doubt our claims of what our tiny 26kb Agent is able to achieve are real, we have provided details below about our patented system producing regenerative machine instructions, Morpheus.

[expand title=”Click here for details about Morpheus”]

Morpheus – Machine Instruction and Data Transformer

Morpheus is a platform that allows Essence tools and applications to run on existing operating systems like MacOS, with Android, ChromeOS, Fuscia, iOS, Linux, tvOS, WatchOS, and Windows versions coming in 2022 and 2023.

Whether applying fixed legacy code (packaged as Powers) or machine instructions generated by the Essence Agent, this Skill brings a powerful capability to efficiently apply transforms (e.g. AI/ML algorithms and more) to any signal or data in real-time, without the need to write code.

Morpheus can transform existing programming jobs into tiny algorithmic units. Examples might include identifying a search pattern, processing a mathematical formula, a file seek and read, and the reordering of data.

Each of these algorithmic units can be expressed in different code templates that produce different machine instructions. Each of these instructions can be bundled and profiled for timing, for energy use, and for resource use and then separated to run on different processor cores based on scheduling access to changes in data.

While some workloads, such as banking transaction processing, are intrinsically serial in nature, the latencies associated with the reading of caches, disk I/O, network packets, and other events can make it possible to split up the work for better performance in many cases.

Some tasks, such as image rendering, sound rendering, and shape generation are inherently optimal for parallel processing. Other tasks such as searching for data patterns, sorting, running mathematical formulas, and logical decisions of a container of data, simulations, and synchronizing precisely timed changes amongst machines can also be parallelized easily. This parallelization can be accomplished using a transform that allows all cores to run simultaneously (or go to sleep if idle) by transforming workloads into algorithmic units and scheduling their instructions across multiple computing processors.

While some tasks inherently have delays, stalls or bottlenecks, the use of tiny algorithmic units maximizes performance with self-profiling and avoids the semaphore, mutex, or locking mechanisms that affect performance in many other parallel systems. This approach to handling work processing requires all work to be done, which is defined as any computational task expressed in the semantic units of Essence Elements, to be estimated for worst, average, and/or best-case duration and resource usage.

The methods used can be mappings between the Essence Elements and instruction blocks, such as: “iterate all elements in Collection A, for each element consider its value B, if it matches C, then add counter D”.

In English, a phrase like: “tell me who I know in Zaire” or “Please show me anyone in my contacts who resides in Zaire”, will map to “Iterate all people in Contacts, Facebook Friends, and LinkedINConnects, and then for each person, if their residence is Zaire, add that person to the collection named ‘People of Zaire’ and then display ‘People of Zaire’ “.

While it is a simple example, it displays each expansion of a basic term into known resources, with a most-recently-stored value for a range (such as the last time we read Facebook Friend List, it was 1000) and iterating per person, using the btree-iterate approach.

So, the machine instructions used is determined for data-access, iteration-of-data, and operations on data (compare it to “Zaire” in this case, which might be GPS-distance or name match or any other method).

These task histories can form an address that can reside in RAM until the space is needed for something else. If RAM space is needed, the data can be cached to disk or dropped and recreated as needed.

These task histories can store the combination of Semantic-Units (the template of activity, such as iterate, compare, find, add) with DataCraft (which databases, which pieces of info used by the Semantic part) and Algorithmic Units.

Algorithmic units identify which actual algorithms and data-reformatting-is needed and was selected, such as using a linear iteration of consecutive addresses (array partition), an incremental pointer de-referencing (a doubly-linked-list), a hash-table, or tree/graph format, etc), which are generally governed by a top-level ‘code-choice’, a midlevel ‘data-format’ (XYZXYZXYZ or XXX YYY ZZZ), and low-level machine instructions (LD, LD, TST, JNE, etc.).

As each task history grows in Data-Craft and Algorithmic Unit histories, the probabilities of making future choices shift based on the accuracy of estimation and the number of optional choices that remain.

For some operations such as square root, there are 2 single-instructions and 4 multi-instruction methods to approximate the value, which is only 6 choices for low precision and only 2 choices for high-precision – a simple case because little variation is possible outside of reordering when the calculation is issued in the task pipeline.

Other cases can be far more complex and have many more expressible choices, which means it may take longer to reach a locally optimal state. Regardless of how much task history data exists, all current tasks are assigned priorities and sorted by resources. We use the classic and effective ‘greedy-solution’ to this NP-complete task, often called the knapsack or traveling salesman problem.

Any delays or missed durations are relayed to the user as required by semantic-unit scope (such as tell me if late, ignore, or log).

It is notable that processor selection alters the Algorithm Unit selection since different instructions may or may not be available as well as accessible ranges of memory usable. It simply results in certain task history scores being set as negative to indicate not applicable.

This approach can have several layers of simulated annealing solutions to the N-tasks using P-processors with I-instructions on R-resources problem. This approach relies on computations being expressed by the Essence Agent.

[/expand]

Introducing Maestro^TM

Learn how we use Maestro to automatically balance resources for best sensory-quality.

[expand title=”Click here for details about Maestro”]

Maestro – Synchronization System

Maestro, a synchronization system that uses an automatic-tuning approach to balancing sensory-quality (such as visual detail or audio fidelity) with responsiveness to user input and latency of changes. Maestro can use a user’s computer language output format (Essence Elements) to detail a series of preferred tradeoffs in any specific sensory art bank or simulation chronicle.

A sensory art bank can be a small rectangular area on a single screen showing a movie or web page or a five-screen wide view of a vast three-dimensional scene or an image on the shirt of a Lego figure inside that scene. For each sensory art bank, there is a tradeoff in how detailed the image or sound or motion or shape (for collision) will be versus how rapidly it responds to changes.

For example, a three-dimensional simulation or high-definition movie that natively generates a new image at 24 times per second (24 Fps) can be synchronized and parallel processed to show a changed visual at 60 times a second (temporal upsampling) or have its pixel density increased to an 8K source (spatial upsampling). If presenting the three-dimensional simulation at 8K requires more processing resources than is available, then the 3D simulation may only be updated at 12 times per second, but the quality will be high, whereas at 60 times per second the quality might be low, but the image is highly interactive.

The synchronization system for parallel processing (Maestro) automates the tuning of controls for visual, audio, physics-simulation, or other sensory and/or calculation based services.

Traditionally, simulations, video games, and general operating systems have controls that computer-savvy users can to tweak to achieve desired results. Applying such preferences to a range of sensory phenomena can become a challenge, especially if the user wants different tradeoffs in different areas, such as crisp but slow-updating text compared to smooth but fast-updated background visuals.

Each element of a video stream (such as a sensory art bank of visuals spread across many screens, many sensory art banks of sounds across speakers, or various physics simulations across space time scenes) can be affected differently. This functionality can be exposed to users as areas-of-focus versus areas-of-less focus.

For example, the window or region of visuals receiving active input can be identified as an area of focus while areas with no immediate user-driven changes can be identified as areas of less focus.

Maestro can use estimates, scores, and course corrections to slowly tune features to keep user response rate, visuals update rate, sound fidelity, physics simulation precision, and general calculation precision in line as close to desired results as possible.

In any scenario, a user can easily overwhelm the available computational resources by simulating too many physical collisions in one time step or by drawing too many details. Maestro can manage this excess demand for computational resources by turning down all tunable tradeoff options and/or pausing the generation of new content until further input is received. This can help preventing stalls or denial-of-service issues caused by the excess demand.

Overall, Maestro can function as a computational governor that helps manage user intent in real time to deliver the optimal experience for a given set of tasks and computational resources. Probabilistic techniques to help determine the cores used, and therefore select where a task is executed. Probabilistic techniques can be used to help determine the instruction-sets initially chosen. The algorithm-units used to generate the actual instructions can rely on probability tables to choose the more likely outcomes. The methods used to assign a likelihood or probability that something may occur, and thus influence decision making, do not need to be coupled with the probability generating methods that produce random values to use in cryptography, visualization, math-solving (such as Monte Carlo solutions), and other approaches.

[/expand]

The Essence Powers System

The Essence Powers System (EPS) has 64 Power Subsystems [8 Categories with 8 Power Subsystems each] that cover all possible ways that an Essence Agent can process native source code for any device, service, computing platform (e.g. drivers, APIs, emulators, browser engines, virtual machines, operating systems, media types, network code, databases, etc.), data, all inputs and outputs.

As Essence has evolved, we have migrated almost all (nearly 80 since December 2019) former Agent Powers into the EPS. Now that the Agent has only 8 Powers, boot/startup times can be instant, making the user experience more like what is expected from an appliance. The much smaller Agent design is easier to customize Essence for operating on existing (e.g. Android, iOS, Linux, Windows, set-top-boxes, TVs, consoles, etc.) and new platforms, including without an operating system.

All Powers in the EPS receive the benefits of dynamically regenerated code (profile and tune for speed, space, bandwidth, wattage, etc.), the ability to upgrade ‘live’…no reboot/restart…(maybe a millisecond delay, but swap in/out fast), flexibility (i.e. run earlier versions along with a new one, when applicable), quicker development times, greater security, as well as lots more benefits.

The Essence Skills System

The Essence® Agent is a powerful way to unlock the potential of machines. Skills empower the individual user to be in control [of the Agent] on an unprecedented level.

As a Developer or a non-Developer, Skills are the ways to access and soon create Powers and Agent generated behaviors without needing to be a coder. Each Skill is presented to the user as an interaction method. Skills can be turned on and off on an App/Aptiv basis.

While other approaches combine code from repositories, our solutions combine both wantware (code generated from expressed intentions) and packaged code (APIs, drivers, emulators, browser engines, media packages and data bases, VMs, and practically any other code) turning them into the digital equivalents of Lego® Bricks. Just plug them in and they work. Hot swap them in and out of memory for much higher efficient use of memory. Our approach makes selecting and combining machine behaviors exceptionally easy and efficient. Skills puts Powers and Essence Agent capabilities in the hands of everyone.