https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/<p style="display:none;">blank</p>en-USTelligent Community 13https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-los/6428967Sat, 25 Jul 2026 02:37:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:d5fb5e0a-4e55-4858-a695-babc25418ee0JanHi Amrit, Thank you for reaching out! [quote userid=&quot;478774&quot; url=&quot;~/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-los&quot;]o Are advertisements transmitted sequentially on advertising channels 37 → 38 → 39? [/quote] The BLE spec requires that advertising channels be used in ascending order, so this ordering is fixed. [quote userid=&quot;478774&quot; url=&quot;~/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-los&quot;]o Is the advertising channel order always fixed? [/quote] Yes, this set by the BLE spec. [quote userid=&quot;478774&quot; url=&quot;~/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-los&quot;]o Is it technically possible to configure the Primary PHY as LE Coded S8 (125 kbps) instead of 1 Mbps? [/quote] Yes, when using extended advertising, you may set the primary PHY to coded. However, the advertisement will only be scannable by coded PHY scanners. [quote userid=&quot;478774&quot; url=&quot;~/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-los&quot;]o Why variation observed in No. of beacon transmitted with above setting e.g. 4-5beacon [/quote] The variation in beacon count (4–5) is expected. The duration value for GAP_ADV_ENABLE_OPTIONS_USE_DURATION is in 10 ms ticks, so 220 maps to 2200 ms. With your primary interval of 800–816 units &#215; 0.625 ms = 500–510 ms, plus the 0–10 ms pseudo-random jitter the controller adds per event, the number of complete events that fit within 2200 ms will vary between 4 and 5 depending on timing. This is normal behavior. Best Regards, Janhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6428889Fri, 24 Jul 2026 21:59:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:8a253659-4ded-425b-ab89-464358d2c863George Mock[quote userid=&quot;601064&quot; url=&quot;~/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6427306&quot;]The only file I couldn&#39;t really find was udivmoddi4.S. [/quote] Here is a comment from near the beginning of that file ... // This file implements the __udivmoddi4 (64-bit unsigned integer divide and // modulus) optimized for code size on ARM architectures. If you are meeting your code size constraints, then using the C implementation should cause no issues. Even if this one detail is different from the TI release. Thanks and regards, -Georgehttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6428874Fri, 24 Jul 2026 21:43:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:6cd0c627-1325-4994-8791-04501f19fe94ZCHi Gullik, I am looking into a 2-GFSK preliminary solution but need to continue working on it into next week (it is taking longer than anticipated). I will post an update with a Studio implementation when it is ready to share. 2 Mbps should be achievable, but as Richard mentions the CC1312/CC1314 will not support higher than this. Regards, Zackhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1667370/cc1352p-packet-loss-observed-at-50meter-losFri, 24 Jul 2026 10:59:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:08df3a95-5980-4a87-8203-265173f7d49fAmrit DhakerPart Number: CC1352P Dear sir, We observed Packet loss observed at 50Meter LOS. With below seeting in transmistter /// Default parameters for long range, connectable, advertising extension #define GAPADV_PARAMS_AE_LONG_RANGE_CONN1 { \ .eventProps = GAP_ADV_PROP_CONNECTABLE, \ .primIntMin = 800, \ .primIntMax = 816, \ .primChanMap = GAP_ADV_CHAN_ALL, \ .peerAddrType = PEER_ADDRTYPE_PUBLIC_OR_PUBLIC_ID, \ .peerAddr = { 0xaa, 0xaa, 0xaa, 0xaa, 0xaa, 0xaa }, \ .filterPolicy = GAP_ADV_WL_POLICY_ANY_REQ, \ .txPower = GAP_ADV_TX_POWER_NO_PREFERENCE, \ .primPhy = GAP_ADV_PRIM_PHY_1_MBPS, \ .secPhy = GAP_ADV_SEC_PHY_CODED_S8, \ .sid = 0 \ } GapAdv_params_t advParamLegacy = GAPADV_PARAMS_AE_LONG_RANGE_CONN1; SendBeacon() { GapScan_disable( ); GapAdv_destroy( advHandleLegacy , GAP_ADV_FREE_OPTION_DONT_FREE); BeaconLen = UdateBeaconAdvertData_g( aBeaconData ); status = GapAdv_create(&amp;SimpleBroadcaster_advCallback, &amp;advParamLegacy, &amp;advHandleLegacy); status = GapAdv_loadByHandle(advHandleLegacy, GAP_ADV_DATA_TYPE_ADV, BeaconLen, aBeaconData); status = GapAdv_enable( advHandleLegacy, GAP_ADV_ENABLE_OPTIONS_USE_DURATION ,220u); /* 2200 ms brodcasting */ } We have below quries. o Are advertisements transmitted sequentially on advertising channels 37 → 38 → 39? o Is the advertising channel order always fixed? o Is it technically possible to configure the Primary PHY as LE Coded S8 (125 kbps) instead of 1 Mbps? o Why variation observed in No. of beacon transmitted with above setting e.g. 4-5beacon Regards, AmritCC1352Phttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6427965Fri, 24 Jul 2026 07:39:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:38fa7d60-7cf8-4ad9-a6cc-bc74f4ca4a9dGullik WebjornUPDATE: Testing with cc1312 revealed it IS possible to reach almost 2 Mbps, with symbol rate 1000 kbps, 166 khz deviation, 1567 khz rx filter. This has also been verified with the cc1314, at 1250 Mhz, 1000 kbps, 166 khz deviation, 1576 khz filter. The RF Studio 7 is somewhat unreliable, frequently when stopping continuous receive / transmit to change parameters, the transfer is not started. If this is a windows issue or a SmartRF Studio issue we have not determined. Suffice to say, if we start it up from scratch, just set the *exact* parameters we want to test, and run both instances, we are rewarded with success, and a 0 % PER running for minutes. This is beginning from the &quot;1st example&quot;, 50 kbps. This resolves the &quot;unable to reach above 1.4 Mbps issue&quot; by using 4FSK @ 1 Mbps. Regards, and thanks to all who helped.... Gullikhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6427306Thu, 23 Jul 2026 18:33:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:a7ca0974-d520-48bf-91a4-d0588e4c2462Daryl OlsenThank you George. I will follow that link. The only file I couldn&#39;t really find was udivmoddi4.S. The only assembly version of that routine in in the llvmorg-15.0.7 tag is in the hexagon/ folder but none exists in the arm/ folder. That said I found udivmoddi4.c version in arm folder. The c version looks to link, compile and work for me so no issue really. Still a bit curious where the assembly version in the libclang_rt.builtins.a sdk release came from but it&#39;s not blocking me.https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6426761Thu, 23 Jul 2026 12:24:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:94144974-73e8-4e80-a6ae-c88d244c86c8Gullik Webjorn&quot;Recommend to follow the register settings used for CC1312R7 1 Mbps, 350KHz Deviation, 2-GFSK, 2.2 MHz RX bandwidth as a base line.&quot; I have done exactly that and eliminated all of OUR settiingparameters. Lenovo G70 -&gt; LAUNCHXL-CC1312R1 on one end ThinkPad i5 --&gt; XDS110-CC1312 other end both with SmartRF Studio 7 NOTE: tested with cc1312 since cc1314 does NOT have a 1 Mbps example, although our target IS cc1314. With the 1 Mbps example: 100 packets sent and received OK, tested several times Modified rate above 1.0 Mbps, by setting the rateword , and deviation = 1/2 symbolrate 1.2 Mbps OK 1.4 Mbps 96/4 packets ok, PER 4%, BER 0,02% 1.5 Mbps running continous mode PER 4.4% 1.6 Mbps continous mode PER 41%, BER 0.29% Tested various bandwidths 1.6 Mbps 3134 khz PER 47%, BER 0.36% 1.6 Mbps 3767 khz PER 47%, BER 0.37% 1.6 Mbps 2486 khz PER 43%, BER 0.29% 1.6 Mbps 2185 khz PER 32%, BER 0.21%, this should require a bandwidth of 800 khz (fundamental of 1.6) + 2 * deviation = 2.4 Mhz, right ?? 1.7 Mbps, failed to get any throughput, after testing at this rate both SmartRF studio required restarts to even run the 1.0 Mbps example, so something screwed up in either the program or the downloaded firmware. What are we doing wrong, should we not be able to reach 2.0 Mbps, as specs claim, or is there some other parameter that needs attending to? BTW, all test performed at reported levels of -40 dBm, so very generous signal. Gullikhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6426647Thu, 23 Jul 2026 10:16:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:c2ac184d-f017-4667-bcd5-10e9654bffaeGullik Webjorn&quot;Recommend to follow the register settings used for CC1312R7 1 Mbps, 350KHz Deviation, 2-GFSK, 2.2 MHz RX bandwidth as a base line.&quot; I have done exactly that and eliminated all of OUR settiingparameters. Lenovo G70 -&gt; LAUNCHXL-CC1312R1 on one end ThinkPad i5 --&gt; XDS110-CC1312 other end both with SmartRF Studio 7 NOTE: tested with cc1312 since cc1314 does NOT have a 1 Mbps example, although our target IS cc1314. With the 1 Mbps example: 100 packets sent and received OK, tested several times Modified rate above 1.0 Mbps, by setting the rateword , and deviation = 1/2 symbolrate 1.2 Mbps OK 1.4 Mbps 96/4 packets ok, PER 4%, BER 0,02% 1.5 Mbps running continous mode PER 4.4% 1.6 Mbps continous mode PER 41%, BER 0.29% Tested various bandwidths 1.6 Mbps 3134 khz PER 47%, BER 0.36% 1.6 Mbps 3767 khz PER 47%, BER 0.37% 1.6 Mbps 2486 khz PER 43%, BER 0.29% 1.6 Mbps 2185 khz PER 32%, BER 0.21%, this should require a bandwidth of 800 khz (fundamental of 1.6) + 2 * deviation = 2.4 Mhz, right ?? 1.7 Mbps, failed to get any throughput, after testing at this rate both SmartRF studio required restarts to even run the 1.0 Mbps example, so something screwed up in either the program or the downloaded firmware. What are we doing wrong, should we not be able to reach 2.0 Mbps, as specs claim, or is there some other parameter that needs attending to? BTW, all test performed at reported levels of -40 dBm, so very generous signal. Gullikhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6425588Wed, 22 Jul 2026 16:31:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:622c6da1-0a66-406c-ba3e-f9127d579bf0George MockYou use the correct branch from the LLVM project on github. Here is one way to tell ... % tiarmclang -c -v empty_file.c 2&gt;&amp;1 | findstr LLVM clang -cc1 version 15.0.7 based upon LLVM 15.0.7 default target arm-ti-none-eabi Even though you use the correct branch, there remains a general concern. Almost all the code used in the tiarmclang compiler is obtained from the LLVM project on github. But some code is changed, other code is added, etc. That said, you are okay in this particular case. Because it is reasonable to need these source files, I filed EXT_EP-13554 to request they be added to the release package. Feel free to follow it with that link. Thanks and regards, -Georgehttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1663993/cc1312r-wireless-m-bus-support-on-cc1312r-for-japanese-920-mhz-band/6424856Wed, 22 Jul 2026 06:44:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:b6fd13c4-de99-46c4-9932-86c682e3b9acIbuki EndoHi expert, Thank you for checking internally. We appreciate your support and look forward to any additional information when available. One additional question: We understand that this may be a difficult request, but if there is any SDK, reference implementation, migration example, or previous work related to Wireless M-Bus operation in the Japanese 920 MHz band, we would greatly appreciate any information you can share. Best regard, Ibuki Endohttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6424177Tue, 21 Jul 2026 18:56:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:e5f5817f-1f91-4999-b1c8-04da8a3e545cDaryl OlsenThank you, I found the needed files at suggested repo.https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661572/cc113l-increase-sensitivity/6423666Tue, 21 Jul 2026 13:29:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:a4899264-583c-4c25-a03e-06be6bb21cbbDaniel Guarecuco AguiarHi Ernest, Have you tried with the exact same configuration as stated in the datasheet (and registers) with the evaluation board? Do you get similar results? Best regards, Danielhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661572/cc113l-increase-sensitivity/6423646Tue, 21 Jul 2026 13:15:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:46cfb06a-a031-447c-b6ef-a4514e2ade3eErnest ChitsonHi Daniel, thank you for the configuration. We have tried it using our datarate / deviation (11500 / 23KHz respectively) configuration, but there was no sensitivity improvement. We would like to try with the evaluation board. What would be the optimal configuration for a sensitivity test using the above datarate and deviation settings ? (78 bits payload length - 94 bits with preamble). Best regards, Ernesthttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-files/6423301Tue, 21 Jul 2026 07:29:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:9c0ad3cf-29e6-47c7-a38d-2b04088b22b0Daniel Guarecuco AguiarHi Daryl, The ticlang compiler is derived from the open source LLVM/Clang source code base which you can find in their Github: https://github.com/llvm/llvm-project/tree/llvmorg-15.0.7/compiler-rt/lib/builtins But this is can be better answered by someone in our Tools/Compiler team. Best regards, Danielhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1660303/cc1312psip-15-4-single-sensor-multi-collector-broadcast-message-support/6422812Mon, 20 Jul 2026 22:01:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:60e8ecfd-f3bf-422b-97a5-9864f57a878aJim GThank you Daniel, this is helpful info for us. Much appreciated!https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1665851/cc1352r-locating-libclang_rt-builtins-a-source-filesMon, 20 Jul 2026 15:01:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:72240e8f-31ea-4fbe-a8ad-82c55178d5f6Daryl OlsenPart Number: CC1352R I am working on a product utilizing CC1352R. For SDK version I used: simplelink_cc13xx_cc26xx_sdk_8_30_01_01 CCS version: 12.8.1 My project is linking in the following llvm clang library that comes bundled with the SDK: ti\ccs1281\ccs\tools\compiler\ti-cgt-armllvm_3.2.2.LTS\lib\clang/15.0.7/lib/armv7em-ti-none-eabihf/libclang_rt.builtins.a In the .map file it is including these files/objects: udivmoddi4.S, aeabi_uldivmod.S, aeabi_memset.S, aeabi_memcpy.S, aeabi_div0.c Due to the certification level of the product, I would like to link these in directly instead of linking the library containing these objects... Can you point me toward the github and/or send the locations of these files?CC1352Rhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6422123Mon, 20 Jul 2026 13:48:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:e27dcdab-6057-4614-a368-3a98f708ae9aGullik WebjornHello RGW, Thanks for feedback. As stated above, we can reach 1.4 Mbps today, which translates to about 0.9 Mbps unidirectional throughput of TCP, which is reasonable. Should I interpret your reply as &quot;4FSK does not work properly at megabit rates&quot; ? We are currently debugging rates 1.5,1.6,...2,0 in 2FSK, and see strange behavior, though our OWN code is still a suspect. Have we reached the limit of capability of the CC1314? The limitations of SmartRF Studio (not being able to set rate above 1 Mbps) has obscured this fact for us..... Gullikhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661347/smartrf-studio-7-4fsk-again/6421912Mon, 20 Jul 2026 09:57:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:b5137282-e23c-4ed0-bd53-484c7f94805cRGWHi, The data rate limitations with CC1314 are 1.5 Mbps to 2 Mbps. You will not be able to achieve a higher data rate than this. I would keep with 2FSK to achieve up to 2 Mbps. Recommend to follow the register settings used for CC1312R7 1 Mbps, 350KHz Deviation, 2-GFSK, 2.2 MHz RX bandwidth as a base line.https://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1664689/cc1312r7-wi-sun-europe-adaptive-power-control/6421850Mon, 20 Jul 2026 08:45:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:50817476-0010-4fdb-b890-055d152bb475Daniel Guarecuco AguiarHi Lucas, You are right, it is partial FAN 1.1 support. Best regards, Danielhttps://e2e.ti.com/support/wireless-connectivity/sub-1-ghz-group/sub-1-ghz/f/sub-1-ghz-forum/1661572/cc113l-increase-sensitivity/6421822Mon, 20 Jul 2026 08:21:00 GMTcb01d8b2-d089-468d-babb-77d1d8683490:eaffa932-6fef-492d-abe5-9e943afc3598Daniel Guarecuco AguiarHi Lilian, 0x21 FREND0 is a valid register, its reset value is already 0x56. You do have it in you configuration #define SMARTRF_SETTING_FREND1 0x56 // FREND1 Front End RX Configuration 0x1E, 0x22, 0x27, 0x28 are not &quot;Reserved&quot; registers, but &quot;Not Used&quot;. Addresses marked as “Not Used” can be part of a burst access and one can write a dummy value to them. Addresses marked as “Reserved” must be configured according to SmartRF Studio ( https://www.ti.com/lit/ds/symlink/cc113l.pdf , Section 5.21) So it is probably fine, but is it something you should test. Best regards, Daniel