Blog Posts

Lately, I've primarily focused on the architectural details of the RX microcontroller (MCU) series that provide performance advantages to design teams. Today let's go back and look at the development-tool issue. Renesas has a new RX610 Starter Kit that packs a formidable set of capabilities into a $250 package. The Kit includes a development board, a Renesas E1 emulator, and the Renesas High-performance Embedded Workshop (HEW) Integrated Development Environment (IDE).

The block diagram pasted below best describes the capabilities of the development board. The design includes an LCD module that is supplied as a detachable daughter card. There is a potentiometer provided for use with the integrated analog-to-digital converters (ADCs). You get a variety of switches and LED indicators for use in evaluation of the architecture.

The kit includes a power supply and all of the cables required to connect the development board to a host PC. You get the option of a RS232 serial connection or linking via USB and the E1 emulator.

We should review a couple of the tools included in the Kit such as the emulator. Renesas offers several options of debuggers and an in-circuit emulator for the RX. The E1 is meant as a tool that's sufficiently affordable that every member of a design team can have their own emulator. The E1 has internal trace capabilities for 256 branches or instruction cycles. The emulator alone sells for $150.

The Kit also includes an evaluation version of the HEW and the Flash Development Toolkit that offer full functionality for 60 days. After the evaluation period the tools still function, but you are limited to the size of projects that can be developed.

In an early post I discussed the capabilities of the HEW and how it's used across Renesas' MCU families. The HEW makes it simple to migrate to the RX from other supported MCUs such as the H8 family. Moreover the IDE includes a complete set of tools such as compilers and a debugger. In addition, many third-party tools integrate seamlessly into the HEW.

I'd also be remiss if I didn't remind you once again that Renesas is still offering the RX-Stick Evaluation Platform for free while supplies last. The RX-Stick lacks the robust features found in the Starter Kit but still provides a chance to evaluate the MCU and the HEW. You can register and request and RX-Stick at the Renesas web site.

Depending on the application’s requirements, different types of LCD configurations can be selected, but basically, we can identify three main types of these LCD configurations that require different driving interfaces:

1.    Simple LCD glass panel without controller/driver.
In this case, an efficient solution is to control the LCD glass using a microcontroller with integrated LCD controller/driver or alternately to emulate the controller/driver using general I/O ports when the number of common and segments lines of the LCD glass is limited. We talk about direct LCD drive for that last solution.

2.    LCD module (or Chip-On-Glass LCD module) driving by parallel interfaces.
Here the module already incorporates the LCD controller/driver in addition of the LCD glass, and it is just a matter of connecting the parallel ports of the LCD module (usually 8-bit for data and 3-bit for control) to I/O ports.

3.    LCD module (or Chip-on-Glass LCD module) driving by serial interfaces, mainly I2C, SPI or RS232.
This is the easiest solution where the LCD module can be controlled by sending commands through the serial lines. Only a few lines are needed to control the LCD module, depending on the serial interface used.

The above classification considers the interface needed to control the LCD, but in each of these configurations, the LCD display can be a segment LCD or a Dot Matrix LCD. Graphic LCD displays also exist but require a more complex driving method.

1. Simple LCD glass panel without controller/driver

To control a simple LCD glass panel that doesn’t integrate a controller/driver, microcontrollers with integrated LCD controller/driver are the best option. The interface consists of connecting the common and segment lines of the LCD to the appropriate pins of the microcontroller. This solution allows saving cost and space on the board as the LCD controller/driver is already provided within the microcontroller chip.
Renesas offers such a solution with the L series’ microcontrollers that integrate such an LCD controller/driver, and are available in 8-bit with the 78K0/Lx3 family and 16-bit with the 78K0R/Lx3 family.

Below is an example of the connection diagram between the 78K0/LF3 MCU and an LCD glass with 8 common lines and 30 segment lines.

The 8-bit 78K0/Lx3 can drive up to 224 LCD while the 16-bit 78K0R/Lx3 can drive up to 400 LCD segments. Also, different display modes can be selected between the following ones:
•    Static
•    1/2 duty (1/2 bias)
•    1/3 duty (1/2 bias)
•    1/3 duty (1/3 bias)
•    1/4 duty (1/3 bias)
•    1/8 duty (1/4 bias)

See It & Sense It! Demonstration kits
Renesas Europe also provides development kits for both of these MCU families, they have been designed to demonstrate and evaluate the 8-bit 78K0/Lx3 and 16-bit 78K0R/Lx3 microcontroller families, but also to support development.

Figure 2 – 78K0/Lx3 See it! and 78K0R/Lx3 Sense it Demonstration kits

More details on these kits can be found here:

http://www2.renesas.eu/products/micro/065_dev_tools/030_starterkits/index.html

When the number of common and segment lines are limited, another possibility to control that type of LCD glass is to emulate the controller/driver using general I/O ports. This method requires emulating the bias voltages using external resistors as shown on the circuit diagram below based on the V850/SA1.

An Application Note related to that direct LCD drive method with LCD controller emulation is available here:

http://www2.renesas.eu/_pdf/U14737EE1V0AN00.PDF

2. LCD module driving by parallel interfaces

An LCD module provides the LCD glass as well as the controller/driver that are already connected together, so that the control is easier for the user. For this type of interface, all microcontrollers can be used assuming that enough I/O ports are available for that purpose, usually eight for data lines and three for command lines (RS, R/W and E) are needed. In addition of the power supply pins, some of those LCD modules also provide two pins for backlight.

The main constraint regarding this type of interface is to use the same operating voltage for both the microcontroller and the LCD module, in other words VDD = VLCD must be fulfilled. But this can be avoided under some conditions allowing the two sides to operate with different supply voltages.
Below is an example of the hardware interface for the 32-bit V850ES/JJ3 and a standard LCD module with an 8-bit parallel interface for data and three more lines for command.

The MCU used in this example, the V850ES/JJ3, as well as all other Renesas V850ES based microcontrollers operating on up to 3.3V, has their I/Os that are tolerant to 5V input. Moreover, assuming that the minimum input voltage (VIH) of the LCD module used is less than 3.3V, therefore, the two sides can operate under different voltages, and no level shifter between the microcontroller and the LCD module is required to adapt the level of the voltages.

Then the appropriate instructions according to the controller/driver datasheet have to be sent to control the LCD.

Renesas provides an Application Note demonstrating this type of interface based on the 32-bit V850ES/Jx3 and Jx3-L MCU families. A software programming example is also available to download as part of the Application Note.

3. LCD module driving by serial interfaces (I2C, SPI or RS232)

This type of LCD module uses the same principle than the previous one except that the control is done using serial interfaces, which reduces the number of lines required compared to a parallel interface.

The connection diagram below shows how to interface an LCD module with an I2C interface to the I2C pins of the Renesas V850ES/JJ3 microcontroller.

The interface only consists of two signal lines for clock and data, and uses pull-up resistors on these two lines.

I've written several times on the subject of interrupt response time and described how the Renesas RX microcontrollers (MCUs) respond to interrupts. It's also important to exit interrupt service routines (ISRs) quickly. A fast exit time allows the MCU to return quickly to the application at hand and also is important when the MCU must service nested interrupts. So today we will take a look at how the fast interrupt feature of the RX and the ability to dedicated registers to an ISR can speed exit times.

The example I'll use here will build on the scenario that I described in a recent post on response to nested interrupts. Renesas application engineers had developed a test in which a timer integrated on the MCU generated an interrupt. Subsequently, a higher-priority external interrupt forces the MCU to exit the timer ISR so that it can service the external interrupt.

The code for the nested interrupt also provided an opportunity to measure the relative advantage of the fast interrupt feature and the use of dedicated registers. I've written previously and described the savings in CPU cycles attributable to the fast-interrupt function and dedicated registers. The fast-interrupt function can be assigned to one high-priority interrupt. It relies on dedicated PC (program counter) and PSW (processor status word) backup registers that eliminate the need to store the RC and PSW to memory. And the dedicated registers eliminate PUSH and POP instructions in the ISR.

The test of how these features speed exit time relied on the use of a scope to measure the time that an I/O signal is in a low state. The timer ISR sets the signal low. And the external interrupt drives the signal back high. The measurement is relative because the timer ISR had to include a delay loop to insure that the external interrupt triggered before the exits from the timer ISR.

First the engineers measured the exit time without the use of the features described above. Next the engineers repeated the test with the timer ISR set to use the fast-interrupt function. The exit required 20.42 nsecs less time with the fast-interrupt function applied to the ISR.

Next the engineers repeated the test with registers R10-R13 dedicated to the timer ISR. That allows the timer ISR to exit without POP operations to restore register values. With both the fast-interrupt function and the dedicated registers applied, the test revealed 80.3 nsecs faster exit time relative to the first test.

Embedded designs often require deterministic and speedy response to interrupts. The RX provides an additional value in lettering design teams cut response times even further for key interrupts.

Some pieces of art are never considered complete; take Gaudi’s Sagrada Familia cathedral in Barcelona – started almost 150 years ago and still under construction despite universal artistic acclaim in the outside world today.

You are probably now wondering if I am really going compare one of the most extravagant pieces of architecture in the world to a microcontroller…

Maybe it’s a step too far, but I believe there are parallels. V850 is acknowledged as the world most popular 32-bit microcontroller, being the world’s top seller for several years in succession now, but as its creators we are still striving to complete the picture; like Gaudi’s cathedral, this picture is not going to be completed any time soon but we are also working to a long-term vision and I would like to share the next step in that vision with you now.

With the introduction of new smaller packages and many more memory options to our existing family, V850 can truly be considered a complete microcontroller platform.

Now spanning from 40-pins up to 144-pins, our V850 J_Series offers a complete platform line-up whilst maintaining flexibility of choice in terms of performance, features such as ultra low power consumption, memory sizes up to 1MB of embedded highly secure Flash, and peripherals such as USB2.0 Host & Function and Ethernet.

Based on a well established 3V high performance microcontroller core, three new flavours have been added to this single family concept:

V850ES/Jx3-L offers market leading power consumption at low voltages, and as well as extending our package line-up, I am pleased to announce that we have also added some new features, including up to 512KB Flash memory, a real time clock with independent power pin and an 11-channel serial array unit.

V850ES/Jx3-U and V850ES/Jx3-H both support 12Mbps Full Speed USB2.0, with either Host and Function or Function only alternatives, plus the added benefit of a CAN interface. We also provide free of charge driver software.

V850ES/Jx3-E offers an IEEE802.3-compliant Ethernet controller as well as USB2.0 and CAN interfaces. To enable designers to take full advantage of all these communication interfaces without restriction, it also includes up to 124K RAM and runs at an impressive 103DMIPS.

And it’s not just pin counts that we’ve considered when introducing smaller packages. We can also offer packages down to 5mm², with standard LQFP plus very low profile WQFN and FBGA packages also offered across the range.

Because we are simply extending a family in full production already, designers are also benefitting from being able to develop early using existing tools whilst the new variants are introduced throughout 2010, so if the one you need isn't quite ready yet, don't panic!

Combined with our existing V850ES/Jx3 family, offering up to 1MB Flash memory embedded, this gives you almost 80 products in a single, upwardly (or downwardly) compatible family, truly enabling you to take advantage of the features you need, be they performance, low power or loads of communications, and break the traditional boundaries of 32-bit microcontrollers with whole new concepts based on one microcontroller platform.

V850 J_Series, now with smaller packages and more features – a complete platform and a piece of art waiting to be held in your hands.

Constant-current drive is a critical part of an LED driver design, since a poorly developed driver can significantly affect the lifetime and color shifting of LED components. Systems that cannot provide a sufficient amount of current will not produce the maximum lighting output, in the same way, driving LEDs with too much current will significantly reduce their duration of service. To use LEDs efficiently, the power source must be able to provide constant current to best match the LED characteristics, where the guidance for optimum currents is provided by LED manufacturers.

World’s First Multi-channel High Current LED Driver MCU

To support the rapidly growing LED market, Renesas offers a unique, highly integrated and intelligent solution for efficiently driving high power LEDs. The uPD78F802x is the world’s first MCU with a 4-channel high current constant driver in a single chip. Each of the channels is capable to drive up to 1.5A using external MOSFETs, which covers the majority of high power LEDs in the market. Additionally, each channel is capable to deliver up to 1MHz switching frequency using an additional internal oscillator, allowing using small size external inductors for each LED channel, thus saving PCB space and system cost. The chip benefits from the wide supply voltage range from 9 to 42V.

Up to 10 high current LEDs can be connected to each channel of the device, which gives a total of 40 LEDs that the device can drive. A designer can leverage the 4 channels for redundancy and drive a high number of LEDs which is beneficial in the lighting applications such as display backlighting or street lighting. And if for one reason or another there is an LED failure in the system, there are 3 other channels which can still take care of that.

To obtain the best match of the light output with the daylight conditions, the dimming capabilities of an LED driver system are becoming essential. Dimming also allows color mixing, which is mandatory in many lighting applications. The uPD78F8024 integrates 4 channels of 8-bit PWM which is capable of generating 256 brightness levels for each channel.

A data sheet for the uPD78F802x family is available here:

http://www2.renesas.eu/applications/industrial/02_building_management/050_lighting/050_led_lighting/020_products/index.html

Shine It! Demonstration kit

The evaluation board included in Shine It! has been designed to demonstrate and evaluate the uPD78F8024 microcontroller and provide a good reference for HB-LED driver system design.

The board is configured in buck topology to drive up to 300mA of constant current for each of the 4 channels.
The board also benefits from flexible LED configurations offering to work with on board 4 tiny RGBW Luxeon Rebel LEDs or utilize the available on board connectors for connecting various external “off the shelf” LED modules.

RS232, DMX512 and ZigBee™ communication interfaces are also supported with relevant connectors allowing remote control of each LED channel independently or as a combination of all the 4 channels. Download the User’s manual for Shine It! here:

http://www2.renesas.eu/docuweb/index.php?fld_keyword=78K0-SHINEIT&fld_issue_date=&fld_type=&submit=Search

Applilet EZ

Shine It! contains the free lighting dedicated Applilet EZ software tool which allows making system configurations using a friendly GUI and automatically generating software for uPD78F802x microcontroller.

It generates pre-compiled IAR code and then program the target device on Shine It! Board using a USB connection. This tool also generates code for DALI and DMX512 lighting specific communications.

The Applilet EZ tool can also be downloaded here:

http://www.renesas.eu/updates?oc=APEZ-HCD

And its user manual is here:

http://www2.renesas.eu/_pdf/U19178EJ5V0UM00.PDF

Introduction

The 78K0/Kx2-L is a low power 8-Bit MCU series designed to target small battery–powered systems and other applications where low power is needed. 78K0/Kx2-L family has onboard OP_AMP with all I/Os available to the user, which makes it possible to implement capacitive touch solution. There are other devices that have the same OP_AMP configuration like the 78K0/Ix2 and 16-Bit 78K0R/Lx3 (-L) which can be found on our website .

Principle of Capacitive Touch

2-1 Theory:
The theory behind capacitive touch can be simply seen from the construction of the two parallel plate capacitor where in free space two conductors separated by a dielectric form a capacitor. Given the fact that the human body is conductive, inspired engineers to use this biological property that uses the effect of body capacitance to alter other conductor’s self capacitance.

2-2 Capacitance Measurement:
When we measure the capacitance in touch system we need to take into consideration the capacitance to ground of the conductive objects. The equivalent circuit can be seen as illustrated in fig-3- where the human finger and the electrode under measurement form a capacitor which lies in parallel to the other capacitors. Therefore the total capacitance is calculated from the following equation:

C1: Body capacitance in order of 100pF
C2: Sensor capacitance in order of few pF
C3: Detection system capacitance to ground in the order of 100pF

The main influence on the measured capacitance comes therefore from C2.

Fig-1-

2-3 Renesas Capacitance measurement Solution
There are many techniques to measure capacitance including direct measurement of the RC time constant, relaxation oscillator capacitance based, switched capacitors, charge transfer and many more. Renesas uses the onboard OPAMP to create astable multi-vibrator with the sensing element capacitance as the tuning element for the oscillator.

The circuit above as the name indicates swings between the rails and therefore the OPAMP output is either Vcc or 0V. The resistor network sets the positive input to the

OPAMP to either 1/3Vcc or 2/3VCC respectively. The capacitor then charges and discharges between 1/3 and 2/3 of Vcc.

The voltage across a capacitor at any time can be formulated as:

By using various functions integrated in the 78K0/Ix2 microcontroller, this device can be used to control high brightness LEDs.  Therefore, the 78K0/Ix2 microcontroller offers a very low cost approach to drive HB-LEDs without an external IC driver.

As part of this low cost approach, the PWM output and internal analog peripherals of the 78K0/Ix2 microcontroller are used for driving constant current and dimming LEDs. These different methods will be explained in details in this post.

1. 78K0/Ix2 Features for HB LED Control

The following 78K0/Ix2 features enables control of high brightness LEDs:

LED constant current drive can be controlled using PWM timers and analog peripherals (comparators and A/D converters). This approach contributes to remove the need of an external LED-dedicated driver IC.

The constant current feedback can be implemented with two different methods:

Comparator feedback

Figure 1 - Block Diagram of 3-channel LED Control using 78K0/Ix2 (feedback by comparators)

10-bit A/D converter feedback

#Figure 2 - Block Diagram of 3-channel LED Control using 78K0/Ix2 (feedback by A/D converter)

LED dimming is controlled using a comparator or PWM timer.

Again, two different LED dimming methods are available for the 78K0/Ix2 microcontroller:

1. DC dimming by changing the internal programmable reference voltage for the comparator

2. PWM dimming by using the 8-bit PWM timer (TMH1)

Three control interfaces for the HB-LEDs are offered using the A/D converter or serial interfaces:

1. Volume control by using the 10-bit A/D converter

2. DMX512 protocol communication control by using UART

3. DALI protocol communication control by using UART in DALI mode

In order to evaluate this low cost control method just introduced, Renesas Electronics Europe offers two demonstration kits, the 78K0/Ix2 LED Control kit which comes with an evaluation board fitted with the 78K0/IB2 to drive three channels of RGB high brightness LEDs, and the Lighting Communications kit for DMX512 and DALI communications, including two easy-to-use GUI tools for each aforementioned lighting protocol.

Figure 3 – New Lighting Demonstration Kits for low cost LED Control

Where can you get more information about these new demonstration kits?

http://www2.renesas.eu/applications/industrial/02_building_management/050_lighting/050_led_lighting/030_resources/index.html

2. Constant Current Control Method

In order to perform control at a constant current, it is necessary to keep the sense voltage through the sense resistors the same as the target reference. The sense voltage is compared to the reference voltage and adjusted accordingly by controlling the PWM output (using an appropriate duty cycle).

The PWM outputs for Power MOSFET switching are generated using the enhanced 16-bit timers X0 and X1. A switching frequency up to 156.25 kHz is available because the timers use a 40 MHz clock  as counting source with an 8-bit resolution (40 MHz / 2⁸).

To perform this comparison, 78K0/Ix2 can use two different methods based either on the internal comparators or on the A/D converter.

2.1 Internal Comparator Feedback

Thanks to its three internal comparators with programmable internal voltage references, the 78K0/Ix2 microcontroller can drive three channels of LEDs independently.

The valid edge of the comparator interrupt has to be specified to be both edges. Then, the comparator interrupt occurs while the sense voltage is higher or lower than the reference voltage. When this interrupt occurs, the constant current control is executed by adjusting the PWM duty cycle, according to the comparator output.

- When sense voltage > reference voltage => Comparator output = “high” => PWM duty cycle is reduced

- When sense voltage < reference voltage => Comparator output = “low” => PWM duty cycle is increased

#Figure 4 – Example of Internal Comparator Feedback

The higher duty and lower duty parameters are stored beforehand.

This method contributes to reduce the software load because the CPU is not accessed before comparator interrupts occur.

2.2 Internal A/D Converter Feedback

The other method employs the 10-bit A/D converter that consists of up to 9 channels so that several LEDs can be driven independently using this feedback method.

The maximum and minimum target values of the A/D conversion results (the reference value for the sense voltage) can be decided based on the detection accuracy of the target current. If the input voltage Vi = V_DD = 5 V, the sense voltage Vs = 1.4 V, and the ratio to define is 2%, the target range of A/D conversion result is supposed to approximately be equal to the target level ±5 LSB.

- When the A/D conversion result is over the maximum while increasing compared with the last result, the PWM duty cycle has to be stepped down.

- On the contrary, when the A/D conversion result becomes smaller than the minimum while decreasing compared with the last result, the PWM duty cycle has to be stepped up.

To speed up the startup period of LEDs, the duty cycle of PWM output has to be set to the target duty value related to the target A/D converter level before starting the timer X and the A/D converter (the target duty related to each target level can be determined by experimenting).

Figure 5 – Example of Internal A/D Converter Feedback

Because converted values are immediately compared, the ripple of current is smaller and the dimming resolution can be higher using this feedback method. However, the CPU is used to perform all channel switching and comparison for A/D converter.

3. Dimming Control

Changing the brightness of LED, which is usually called dimming, can be achieved by varying the LED forward current. To do so with the 78K0/Ix2 microcontroller, two dimming methods are available: DC dimming and PWM dimming.

3.1 DC Dimming

By changing the internal reference voltage

As expressed earlier, when using the comparator feedback method, the 78K0/Ix2 microcontroller monitors the sense voltage and compares it to a stable reference voltage to keep the current constant. Therefore, changing this reference voltage contributes to modify the constant current to a different level, and the brightness of the LEDs reacts accordingly.

In that case, the internal reference voltage of each comparator has to be changed, as well as the higher duty and lower duty parameters.

The internal reference voltage can be changed in 32 steps in the range from 0.05 V (typ.) to 1.6 V (typ.). To increase the dimming resolution, the sense voltage when driving current at the maximum brightness must be set to as high a value as possible in the range of the internal reference voltage.

By changing the target level for the A/D conversion  result

For the A/D converter feedback method, dimming consists in changing the target level, as well as the target duty.

The reference voltage for the maximum brightness is selected by considering the hardware design and the limit of the A/D converter reference voltage. The dimming step number is decided by using maximum reference voltage and error margin for the target level.

To increase the dimming resolution, the sense voltage when driving current at the maximum brightness must be set to as high a value as possible within the range of voltages for which A/D conversion is possible (AVREF or less).

3.2 PWM Dimming

The 78K0/Ix2 microcontroller includes a function that performs gate control for the signals output by the 16-bit timers X0 and X1 (TMXn) by using the output of the 8-bit timer H1 (TMH1).

Keeping the reference voltage at the maximum level, LEDs can be dimmed in 255 steps by using the TMXn output gate function via the TMH1 output. A square PWM wave output by the 8-bit timer H1 can be set to control all four TMXn outputs. If TOH1 is combined with the outputs of 16-bit timers X0 and X1, the TMXn output is only enabled when the TOH1 output is at high level or low level.

Figure 6 – Timing Chart for PWM Dimming

While using PWM dimming via TMH1, all TMX outputs will start or stop at the same time. Therefore, this method can be used to dim the LEDs of all channels at the same time.

Where can you get the related Application Note?

http://www2.renesas.eu/_pdf/U19666EJ1V0AN00.PDF

Author: Philippe Miquel, Graduate Engineer, Renesas Electronics Europe.

Today I'd like to discuss interrupt response again relative to the Renesas RX microcontroller (MCU) architecture. We'll add an external interrupt into the mix and prepare for a later examination of how to minimize the exit time for interrupts.

Let's first take a look at the type of response that you can get from nested interrupts. We'll rely on the scenario that I described in the first post that I did on measuring interrupt response time. The test described measuring the interrupt response when an integrated timer was used to generate an interrupt upon overflow. The interrupt service routine (ISR) then read the value of the timer, and that value was the measure of the interrupt response time because the timer resets itself to zero upon overflow in addition to generating an interrupt.

Application engineers changed the test code bringing an external interrupt into the equation. The new test relies on the timer ISR to drive a signal on the RX610 prototype board low that in turn triggers an external interrupt on IRQ14. In the test code, IRQ14 was configured as a falling-edge interrupt with a priority that's higher than that of the timer interrupt. So the external interrupt causes the RX to stop the timer ISR and service IRQ14.

The IRQ14 ISR does little more than drive the triggering signal back to a high state. The application engineers used a scope to measure the time the triggering signal was low, to measure the response to the external IRQ14 interrupt. And remember that servicing that interrupt required that the RX first exit the timer ISR.

The external interrupt response was measured to be 320 nsecs or 32 CPU clock cycles. Of that 32 cycles, 6 cycles were spent in the IRQ14 ISR. So the time it took to exit the timer ISR and dispatch the first instruction in the IRQ14 ISR is 26 CPU cycles.

We can take a brief look at what happened during the 26 CPU cycles. The RX must POP the registers used by the timer ISR and restore the program counter (PC) and processor status word (PSW). The interrupt control unit (ICU) must process the new external interrupt, load a new interrupt vector from memory, and transfer control to the new address.

I've described the nested interrupt scenario for a couple of reasons. First the 26-cycle response is really good for servicing a nested interrupt. Moreover it could be made faster using the same techniques that I described in the post on the fast interrupt function and dedicated interrupt registers. If IRQ14 was deemed the most critical interrupt in the system, the backup PC and PSW could have shaved a couple of clock ticks off the interrupt response time, and dedicating registers to the ISR could save additional cycles in the IRQ14 ISR.

I also described this example to set the stage for another post on how the fast interrupt function accelerates the exit from an ISR – coming soon.