espressif's "Chip7 2-2-A" spotted in the wild

WiFive
Posts: 3529
Joined: Tue Dec 01, 2015 7:35 am

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby WiFive » Thu May 23, 2019 11:36 am

WiFive wrote:
Mon Mar 25, 2019 6:27 am
Sounds like it is basically "esp32+" or "esp33". I think we can mention small things that are hopefully fixed/upgraded from esp32 since it is not in production yet, there is still time!

- More psram cache address space
- More flash cache address space
- better ulp core (some tiny xtensa thing?, c targetable)
- minimal iomux conflicts
- better i2c peripheral, support clock stretching, slave address mask, monitor/debug mode
- support espnow and ble advertising from rom/ram only (wake stub, no flash load)
- BLE 5 phy features
- better continuity of reserved ram areas
Review the wish list...
- More psram cache address space ✓
- More flash cache address space ✓
- better ulp core ✓
- minimal iomux conflicts ✓ (we'll assume more pins means yes)
- better i2c peripheral, support clock stretching, slave address mask, monitor/debug mode :?: (maybe)
- support espnow and ble advertising from rom/ram only (wake stub, no flash load) ✖
- BLE 5 phy features ✖
- better continuity of reserved ram areas ✓ (probably)

ESP_Sprite
Posts: 9766
Joined: Thu Nov 26, 2015 4:08 am

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby ESP_Sprite » Fri May 24, 2019 3:33 am

I think the I2C slave implements some kind of clock stretching now, so there's at least half a ✓ more :P

User avatar
Vader_Mester
Posts: 300
Joined: Tue Dec 05, 2017 8:28 pm
Location: Hungary
Contact:

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby Vader_Mester » Fri May 24, 2019 5:08 am

ESP_Sprite wrote:
Fri May 24, 2019 3:33 am
I think the I2C slave implements some kind of clock stretching now, so there's at least half a ✓ more :P

Will it be able to load and run programs from PSRAM?

Code: Select all

task_t coffeeTask()
{
	while(atWork){
		if(!xStreamBufferIsEmpty(mug)){
			coffeeDrink(mug);
		} else {
			xTaskCreate(sBrew, "brew", 9000, &mug, 1, NULL);
			xSemaphoreTake(sCoffeeRdy, portMAX_DELAY);
		}
	}
	vTaskDelete(NULL);
}

ESP_Sprite
Posts: 9766
Joined: Thu Nov 26, 2015 4:08 am

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby ESP_Sprite » Fri May 24, 2019 6:06 am

I think psram also has I-port access now, so yes.

User avatar
rudi ;-)
Posts: 1731
Joined: Fri Nov 13, 2015 3:25 pm

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby rudi ;-) » Fri May 24, 2019 6:22 am

ESP_Sprite wrote:
Fri May 24, 2019 6:06 am
I think psram also has I-port access now, so yes.
incredible moment
if that's possible - who should be able to top that setup?
espressifer's digi guys - keep up the very good work !!
the last 5 years we have invested well, I do not regret a minute.
Excellent!

best wishes
rudi ;-)
-------------------------------------
love it, change it or leave it.
-------------------------------------
問候飛出去的朋友遍全球魯迪

User avatar
rudi ;-)
Posts: 1731
Joined: Fri Nov 13, 2015 3:25 pm

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby rudi ;-) » Fri May 24, 2019 7:11 am

few things are spotted in the wild last hour's..

Engineering Board PIC taken from the ESP32-S2 demo 1 , also here
ESP32-S2_Demo1.jpg
ESP32-S2_Demo1.jpg (34.37 KiB) Viewed 13087 times
best wishes
rudi ;-)
-------------------------------------
love it, change it or leave it.
-------------------------------------
問候飛出去的朋友遍全球魯迪

Deouss
Posts: 425
Joined: Tue Mar 20, 2018 11:36 am

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby Deouss » Fri May 24, 2019 1:40 pm

May I ask - why not 2 cores?

:?:

ESP_Sprite
Posts: 9766
Joined: Thu Nov 26, 2015 4:08 am

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby ESP_Sprite » Sat May 25, 2019 11:37 am

Because cheaper. We're still of the opinion multiple cores is a good idea, just not in this specific chip.

User avatar
Vader_Mester
Posts: 300
Joined: Tue Dec 05, 2017 8:28 pm
Location: Hungary
Contact:

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby Vader_Mester » Sun May 26, 2019 5:24 pm

Well, 240MHz is well enough. Although, in this case FreeRTOS could be excluded entirely, and maybe, the chip's resources could be used more efficiently.

However I still think that in more complex cases FreeRTOS or something simillar can still be a good option.

One request for the devs: Please make a WROVER style module from the start. Also - I guess I probably ask too much - but make an IDF config option to be able to configure the chip to be able to correctly run code from PSRAM.

I can strongly predict, that as soon as this is launched, someone will make an OS for the chip, and not long after, someone will port a Linux onto this thing - thanks to the PSRAM instruction running capability, and 128MB PSRAM support! :)

Code: Select all

task_t coffeeTask()
{
	while(atWork){
		if(!xStreamBufferIsEmpty(mug)){
			coffeeDrink(mug);
		} else {
			xTaskCreate(sBrew, "brew", 9000, &mug, 1, NULL);
			xSemaphoreTake(sCoffeeRdy, portMAX_DELAY);
		}
	}
	vTaskDelete(NULL);
}

User avatar
rudi ;-)
Posts: 1731
Joined: Fri Nov 13, 2015 3:25 pm

Re: espressif's "Chip7 2-2-A" spotted in the wild

Postby rudi ;-) » Tue May 28, 2019 7:00 am

ESP_Sprite wrote:
Sat May 25, 2019 11:37 am
Because cheaper. We're still of the opinion multiple cores is a good idea, just not in this specific chip.
mhm one core, dual core,
what is with ( pspspspspspsp ) ESP32-S3x
how much cores?
can you wisper a little? ( only little, come on :?: )
:o

not that i did not say it - but have everyone noted that the NEW is not LX6 but it is LX7 ?
2 FLOPS / cycle for the LX6, but 64 FLOPS / cycle for the LX7 ( Info ENG , Info DE )

hosa :P

espressif's own DSP is comming :o :shock: :)

exciting year!

best wishes
rudi ;-)
-------------------------------------
love it, change it or leave it.
-------------------------------------
問候飛出去的朋友遍全球魯迪

Who is online

Users browsing this forum: Majestic-12 [Bot] and 86 guests