2024-08-09 04:33 AM
Hi,
The datasheet DS13311 rev. 5 defines the max. SPI clock for the OCTOSPI only for VOS0, what is the max. SPI clock for other Power Scale settings (I'm planing to use VOS1)?
best regards
martin
Solved! Go to Solution.
2024-08-14 07:27 AM
To close this:
I have got an answer to my case from ST:
In you case at VOS1, the performance can be degraded by up to 5% compared to VOS0.
Case closed
Martin
2024-08-09 05:48 AM
Just found this in my source code, reference to RM0468, page 336,
and I'm using a HyperRam at 100 MHz on H733:
/* notes:
* Power:
* - voltage scaling for core:
* VOS0 ~1.36 V - "boosted performance" - highest freq -> highest power
* VOS1 ~1.21 V - "high performance"
* VOS2 ~1.10 V - "medium performance"
* -> in these voltage ranges SAI, I2S, OctoSPI can work with at least up to
* 113 MHz (RM0468, pg 336)
*/
void SystemClock_Config(void)
{
RCC_OscInitTypeDef RCC_OscInitStruct = { 0 };
RCC_ClkInitTypeDef RCC_ClkInitStruct = { 0 };
/* supply configuration update enable */
/* no SMPS on H733 */
HAL_PWREx_ConfigSupply(PWR_LDO_SUPPLY);
/* configure the main internal regulator output voltage */
__HAL_PWR_VOLTAGESCALING_CONFIG(PWR_REGULATOR_VOLTAGE_SCALE1);
while( 0 == __HAL_PWR_GET_FLAG(PWR_FLAG_VOSRDY) );
...
2024-08-09 07:05 AM
Hi LCE
Nice finding, i would never have search in the reference manual!
So according to this the max. Pclk for the OCTOSPI with VOS1 is 200MHz, but there is no information about the max. SPI clock freqency. Thanks for the information that You have a HyperBus device running at 100MHz, but I whould like a confirm from ST that this is OK, probably I need to open a case.
Martin
2024-08-09 08:13 AM - edited 2024-08-09 08:13 AM
Hello @MHoll.2 ,
The STM32H725 datasheet cited only the max OCTOSPI frequency with VOS0.
As VOS1 is lower than with VOS0, the maximum OCTOSPI with VOS1 will be lower than the maximum OCTOSPI with VOS0.
Thank you.
Kaouthar
To give better visibility on the answered topics, please click on Accept as Solution on the reply which solved your issue or answered your question.
2024-08-10 01:46 AM
Hi Kaouthar,
Thanks for You answer, I was expecting this, but can You (or someone in ST) tell me how much lower?
This is not some thing I can test my self easily, I don't won't to have to recall a product because 1% of the products are not working stable, but on the other hand I need as much performence from the two OCTOSPI as possible (they will be used for Framebuffer and Code)
Martin
2024-08-14 07:27 AM
To close this:
I have got an answer to my case from ST:
In you case at VOS1, the performance can be degraded by up to 5% compared to VOS0.
Case closed
Martin
2024-08-14 07:52 AM
Dear @MHoll.2 ,
Thank you very much to be back to our Community with the answer from our colleagues in charge of the R&D :
As it is the using the same technology with STM32H7A/B we can apply the same
and it would be an opportunity to update the H725/35 Datasheets .
Ciao
STOne-32.